Coding agents, skills, prompting craft and agent-native workflows.
147 fiches · 243 entities · Updated
Coding agents are software systems that plan, write, and revise code through iterative model-driven loops rather than single completions. Gathered here are analyses of how they work in practice — the prompting craft that steers them, the skills and tools that extend them, and agent-native patterns like plan mode, subagents, and harnesses. Recurring through the fiches: what makes an agent dependable on a real repository, how the unit of engineering slides from the keystroke to the loop, and where autonomous coding still fails. Named practices — vibe coding, augmented coding, loop engineering — appear as the field labels what it is learning. The material favors reported practice and failure modes over vendor promises.
Essay by **Bill Staples**, CEO of **GitLab**, published on **August 24, 2026** on the about.gitlab.com blog: an announced **31-minute** read, roughly **39,000 characters**, presented as the sequel to a memo written to the board of directors in January 2026 and partially published in May under the title *GitLab Act 2*. The text presents itself as a response to **Anthropic**'s AI-native SDLC playbook, published three days earlier, from which it borrows the opening line — "Code is no longer the bottleneck" — to pose the question that drives it: what becomes scarce when code becomes abundant. (A) The economic diagnosis: the useful unit is not cost per line but **cost per accepted change**, which aggregates generation, environment, context, verification, review, remediation, and governance; AI collapses the generation term alone, which makes the others proportionally heavier — an organization ten times faster at generating "will simply move the queue." (B) The architectural response: four capabilities — agent platform, machine-scale execution, durable context, governance — forming an enterprise layer that outlives the model, "The model should be replaceable. The agent should belong to the customer." (1) Three modes coexist durably, from human-driven legacy to autonomous development, against the idea of a single maturity curve. (2) The CI/CD pipeline becomes the place where the inner loop runs, instead of being an end-of-line gate. The figures cited are Stripe's, Spotify's, and Amplitude's; GitLab produces only one, about its own source control. The corpus already holds [[claxton-anthropic-ai-native-sdlc-playbook-2026-08-21]], the source this text responds to, and [[sfeir-sdlc-pdlc-articulation-2026-07-22]] on the SDLC/PDLC articulation that Staples adopts as his own.
#code abundance#cost per accepted change#theory of constraints
Bill Staples · directeur général de GitLab (fonction non affichée par la page) · sur le blog about.gitlab.com.
Long-form guide from **Anthropic** by **Louis Claxton** (Applied AI team), published on **August 21, 2026** on the claude.com blog: a stated **40-minute** read, roughly **64,000 characters**, presented as a collection of *plays* drawn from the team's work with its clients. (A) The diagnosis: with code no longer the bottleneck, it shifts to the stages on either side of the build (plan, review/test, deploy), line-by-line controls stop holding once the agent writes most of the diff, and governance cost rises as exceptions still route through periodic committees. (B) The response: six stages (Plan, Design, Build, Test, Deploy, Maintain) organized as a **loop** rather than a chain, each ending with a **committed artifact** that the next stage reads — `intent.md`, `spec.md`, `plan.md`, the diff and its tests, the PR and its findings, the incident record. (1) Institutional knowledge becomes versioned files: `CLAUDE.md`, skills, `REVIEW.md`, `bands.yaml`. (2) Governance splits into two layers, with the skill positioned as an advisory control and the hook as the deterministic layer behind it. Separation of duties is set as an invariant — the agent that writes the code cannot approve it — and the piece closes on *"The loop keeps running. Human judgement stays above it."* The corpus already holds [[clinton-anthropic-secure-ai-native-sdlc-2026-07-21]] on the security side of the same cycle, and [[hingel-augment-how-ai-changes-sdlc-six-stages-2026-06-08]] on the same six-stage breakdown as seen by a competitor.
#AI-native SDLC#software development lifecycle#plays
Louis Claxton (Anthropic, équipe Applied AI) · sur le blog claude.com ; contributions créditées à Jim Blackhurst · Will Steuk et Jamal Arif.
Guide signed by **Michael Segner**, published on **August 20, 2026** on the claude.com blog in the *Claude Code* category: a **5-minute** read announced for approximately **31,500 characters** of body text, also offered as a PDF. Stated material: interviews with **more than a dozen** startups, fifteen named — **Artemis Security**, **Cainex**, **Clay**, **ClickHouse**, **Cognition**, **Commure**, **Crosby**, **Emergent**, **Harvey**, **Heidi**, **Higgsfield**, **Omni**, **Parahelp**, **Translucent**, **Zingage**. (A) Five operating rules: *everyone ships*, *automate the tedium*, *trust, but verify*, *build for rebuilding*, *prototype, dogfood, productionize*, each closed with product tips and gathered into a final checklist. (B) A body made of attributed quotes, each rule illustrated by named executives rather than by an aggregated metric. The four figures highlighted are those of the interviewed companies: **+30%** more features shipped (ClickHouse), **2 to 3×** engineering productivity (Omni), **100%** of bug triage automated (Clay), **more than 6,000 PRs per week** (Artemis Security). Two passages depart from the testimonial register: **Cainex**'s self-correction loop on medical coding, described step by step, and the internal use of **Claude Tag** at **Anthropic** as first responder for CI/CD on-call. The question posed at the opening — *"what would it look like if an organization built their product development lifecycle with Claude Code from the ground up?"* — connects with [[claxton-anthropic-ai-native-sdlc-playbook-2026-08-21]], published the next day by the same publisher, and extends [[cherny-wu-reflecting-year-claude-code-2026-07-17]].
#Claude Code#startups#everyone ships
Michael Segner · auteur du guide sur le blog claude.com (fonction non affichée par la page) ; entretiens avec les dirigeants de quinze entreprises nommées.
Corporate blog post from **Block** (`block.xyz/inside`), unsigned — the displayed author is **"Block"** —, published on **August 18, 2026**, ~930 words, announcing **the open-sourcing of Berd**, Block's internal desktop application for working with agents, and laying out the design thesis that guided it: giving agents character *"not only through roles, instructions, skills, and tools, but through distinctive visual identities"* — hence the in-house animated characters, the *"Gloopies"*. The post starts from an observation of fragmentation (*"The technology was powerful, but the experience around it was fragmented"*) and a precisely named interface problem: *"the product gives people little sense of how the agent is configured, which context and tools are available to it, and how it differs from another agent"*. Two structuring contributions. **(A) A three-tier articulation**: **goose** remains the framework and *runtime* that holds the agent loop; **Berd** is the desktop client (projects, context, sessions, agents, configuration); the two communicate via the **Agent Client Protocol**. **Buzz** is designated as the follow-up, for when solo work becomes collaborative (*"Start alone, then go multiplayer"*). **(B) Six requirements handed off to Buzz**, stated as a takeaway: *"private space, durable context, recognizable agent identities, reusable skills, visible configuration, and clearer visibility into an agent's configured context, tools, and capabilities"* — a grid directly reusable for evaluating an agent client. The text itself distinguishes identity from capability: *"The avatars make the agent recognizable. Its role, skills, and tools make it useful."* No usage figures are produced and no license is named for the open-sourcing.
#Berd#Block#open source
**Aucun auteur nommé** : le billet est signé **« Block »** — le champ *Author* de la page porte le nom de l'entreprise. Publié le **18 août 2026** sur `block.xyz/inside` · le blog **corporate** · et non sur `engineering.block.xyz`.
Blog post from **Sonatype** by **Aaron Linskens** (*technical writer*), published on **August 18, 2026**, ~1,300 words: it recounts a **Sonatype Research Labs** study spanning **49 months** (June 2022 — June 2026) and a **fixed cohort** of enterprise applications, a methodological choice asserted to isolate the evolution of the application fleet rather than that of the customer portfolio. The result is presented as a contradiction: remediation is faster, yet risk accumulates further. (A) **The stock is rising** — *Critical* and *High* vulnerabilities per application **×4.31** (from **14.14** in June 2022 to **54.3** in 2026, still **×3.91** excluding legacy applications newly brought under management), newly affected component versions at **46×** the pre-AI rate, monthly application creation **×4.84**. (B) **Remediation is improving** — more than half of resolved violations are resolved in under a day, the median age of unresolved *Critical/High* vulnerabilities drops from **228** to **126 days**, then to **103** in May 2026; among cohorts that had twelve months, **52.6%** are resolved, **44.3%** open, **3.1%** under waiver. (C) **The proposed lever is component selection**: at the moment a vulnerable dependency was chosen, a substantially less risky version already existed in **62.2%** of cases on **Maven**, **46.9%** on **npm**, **34.3%** on **PyPI** — a gap the text attributes to an information gap rather than developer fault. The post itself states that AI is not the sole cause of the acceleration, and concludes on **Sonatype Guide**, which brings this intelligence to the point of selection. On the supply-chain side, it extends what [[fiches/2026-08/staples-gitlab-when-code-is-abundant-2026-08-24]] frames in economic terms and [[fiches/2026-07/clinton-anthropic-secure-ai-native-sdlc-2026-07-21]] in secure-cycle terms.
#software supply chain#software supply chain#Sonatype Research Labs
Aaron Linskens · *technical writer* chez Sonatype · sur le blog de l'éditeur ; les chiffres sont produits par Sonatype Research Labs · non par l'auteur.
Product announcement post from **Block Engineering** signed by **Thomas Petersen** (*Principal Designer & Builder*), published on **August 18, 2026**, ~1,800 words across thirteen short sections, introducing **Buzz Projects** — a **software forge hosted on its own relay**: Git repositories, branches, pull requests, issues, review and merge, multi-repo projects, an activity feed, all linked to conversation channels. The post's standfirst and thesis: *« Coding agents are the terminal for your computer. Buzz is the terminal for your network. »* Three contributions. **(A) A trust doctrine grounded in *ex post* proof rather than *ex ante* authorization**: on one side *« No forced guardrails, no limitations on what your agents are allowed to help you with »*, on the other *« Every push, review, approval, and merge is a signed Nostr event. If an agent authors a patch, you can see which agent produced it and which human authorized that agent to act »*; the section closes on a stated direction — *« we are already exploring ideas around agent trust protocols informed by past behavior »*. **(B) Git interoperability without proprietary tooling**: *« These are standard git repositories… You can fetch, clone, pull, and push over plain Smart HTTP, with no custom tooling or wrapper CLI required »*, with the clé Nostr serving as a single identity — *« The same npub that signs your messages signs your pushes. »* **(C) A distinction between execution surface and network presence**: *« A terminal gives an agent somewhere to execute commands and change files, but it does not give it a persistent place in the network. Buzz does. »* The post produces no figures and contains no outbound links; it qualifies itself as preliminary six times (*« still very basic »*, *« fairly elementary »*, *« still under experiments »*), and Projects lives under the **Experiments** tab of Buzz Desktop.
#Buzz#Buzz Projects#Block
**Thomas Petersen** — *« Principal Designer & Builder »* chez **Block** · auteur unique et signataire du billet ; première apparition dans le corpus. Publié le **18 août 2026** sur le blog **Block Engineering**. Troisième signature Block sur Buzz en un mois · après Tyler Longwell (21 juillet) et Atish Patel (6 août) · et la première non-ingénieur.
X post by **Andrew Ng** from **August 14, 2026** (16:29 UTC), reprising the "Dear friends" letter from ***The Batch* #366** (DeepLearning.AI, same date), ~900 words. Ng presents **The AI Engineering Skills Map** and publishes **four skills** held to be the most important. **(1) Building and deploying AI applications** — the specificity is named: *« The key difference between AI and non-AI applications is that the former has unpredictable outputs »*, hence the emphasis on *evals* and error-analysis loops. **(2) Software engineering fundamentals**, because *« Understanding software fundamentals allows you to recognize what tradeoffs even exist »* — the inexperienced developer fails *« because they don't know what context to give their coding agent »*, hence the goal of *« steering coding agents using the precise language of software engineering »*. **(3) Using coding agents**, in an operational formulation: *« help the agent autonomously close loops by providing verifiers or evals »*, and *« knowing how much to intervene and how much to leave them alone »*. **(4) *Shaping the build***: *« Given a clear spec, coding agents are rapidly improving at delivering to it. Thus, our work as engineers is shifting toward deciding what should be in the spec »*, paired with *« Engineers should no longer expect to be given a pixel-perfect design and asked only to implement it. »* A **terminology note** carries most of the framing: Ng talks about **skills** in AI engineering and **not the role** "AI Engineer", with an explicit analogy — *« All developers today should know how to work with the cloud, and only a smaller number have a "Cloud engineer" title. »* The whole is backed by *« an analysis of more than 10,000 job postings, dozens of structured interviews with experts, hiring managers, and recruiters, surveys, and other online data »*, of which **no numeric results are published**: Ng describes his process as *« informally… akin to running clustering »* and announces a detailed map in future posts. He states the interest in the second-to-last sentence: *« DeepLearning.AI's principal focus is to help developers gain these AI engineering skills. »*
#AI Engineering Skills Map#skills map#Andrew Ng
**Andrew Ng** — fondateur de **DeepLearning.AI** · general partner d'**AI Fund** · cofondateur de **Coursera** et de **Google Brain** · ancien chief scientist de Baidu. Texte signé · à la première personne · écrit *« with my team »* sans qu'aucun collaborateur soit nommé. Publié le **14 août 2026** sur X et dans ***The Batch* n°366** — même texte aux deux endroits ; préférer *The Batch* pour toute citation durable. Quatrième fiche Ng du corpus · après les lettres n°350 (24 avril) · n°352 (8 mai) et n°359 (26 juin).
Announcement post published on the **official Z.ai blog** (formerly Zhipu AI, Chinese lab) on **August 14, 2026**, **with no individual byline**, ~2,000 words plus footnotes. It announces **GLM-5.3**, successor to GLM-5.2, opening with a methodological thesis: *« Scaling post-training is all we did for GLM-5.3. »* Same base model as GLM-5.2 — *« every gain comes from post-training »*. Three announcements. **(A) An open-weights coding model**: +50% claimed on **Z.ai Code Bench**, an unpublished in-house benchmark. **(B) A cyber capability presented as "emergent"**, which the body of the text traces to a training choice — *« As part of post-training, we introduced vulnerability discovery data and environments into the training mix. We expected this to make the model better at finding and reasoning about vulnerabilities »* — what came as a surprise was the speed and the change in nature: the model moves from identifying isolated flaws to *« coherent plans for complete exploitation chains »*. Gains grow with position in the exploitation chain: CyberGym 77.2 → **84.5%**, ExploitBench 24.4 → **54.4%** (×2.2), ExploitGym 29 → **105** tasks in 2h (×3.6), with the gap to the closed frontier remaining wide (181 and 247 tasks). Z.ai puts it this way: *« Capability is growing fastest exactly where we are furthest behind. »* The post also publishes a **Z.ai Security Disclosure Ledger**: **2,436 vulnerabilities identified across 269 open source projects** — kernels, OSes, browser engines, infrastructure, web applications, network protocols — the oldest introduced in **1981**, average lifetime before discovery **26.6 years**, of which **53 disclosed** and **2,383 under embargo**. **(C) A weight release** *« within two weeks of launch, once safety evaluation and hardening are complete »*. The most reusable methodological contribution: **environment and verifier synthesis**, the latter produced without access to the reference solution and admitted only after a triptych of negative controls — **oracle**, **no-op**, **unsolved-state**. All agentic evaluations are conducted **in Claude Code 2.1.207**.
#GLM-5.3#GLM-5.2#Z.ai
**Z.ai** (anciennement **Zhipu AI**) · laboratoire d'IA chinois · éditeur de la famille **GLM**. Billet **institutionnel et non signé** : aucun auteur nommé · aucun chercheur mis en avant · aucun lien vers un rapport technique ou une carte de modèle. Publié le **14 août 2026**. La page est une SPA React — le HTML servi est un `<div id="root">` vide · et le texte comme les scores ont dû être extraits du bundle `glm-5.3-BCnx8T5_.js` · où ils figurent en valeurs source.
Official product page from **DeepSeek**, published on **August 13, 2026**, **unsigned**, ~450 words, announcing the *developer preview* release of **DeepSeek Harness** (`dsh`) — a coding-agent harness **open source under the MIT license**, whose repository opened the same day. A three-word thesis, repeated in the title and in the repository description: *« Everything is a plugin »*, paired with a second promise, *« Every run is traceable »*. The page states the equation *« AGENT = MODEL + HARNESS »* and lists the pluggable capabilities — *« models, tools, skills, sessions, sandboxes, storage, loops, scheduling, and the UI »*. Four modes ship: **Standard** (full coding agent), **Code** (tools exposed via the *Code Mode SDK*, letting the model compose multi-step operations inside a TypeScript program), **Minimal** (*« two-tool coding agent with persistent bash and str_replace_editor »*, explicitly *« for benchmarking models in a minimal environment »*), and **Creator** (runtime inspection, in-memory plugin testing). The technical substance sits in the repository, not on the page: `docs/architecture.md` states a logging invariant — *« Model-visible means logged. Anything that reaches a model request must be reconstructable from the log, and a runtime invariant asserts it »* — and states that *« there is no privileged core to patch »*. The technical core is not DeepSeek's own: DSH is built on **Cordis** (the `cordiverse` project, a third party), **vendored** into `vendor/` with a manifest and a sync procedure, and the page places the *« Cordis paper »* at the same navigation level as "GitHub" and "Developer docs". Two LLM adapters ship — `dsh-llm-deepseek` and `dsh-llm-pi-ai`, a generic multi-provider adapter. The repository warns in capitals: *« THERE WILL BE COMPATIBILITY-BREAKING CHANGES »*, and `CLAUDE.md` specifies that `SESSION_FORMAT_VERSION` stays at `0` *« with no compatibility promise »*, with backends rejecting old on-disk formats. Timeline: DSH ships on the day **DeepSeek-V4-Pro reaches GA**, three days before a new API pricing schedule takes effect on **August 16, 2026 at 16:00 UTC**, with peak/off-peak rates and an off-peak discount of **−50%**.
#DeepSeek Harness#dsh#agent harness
**DeepSeek** (DeepSeek AI, laboratoire chinois) · en tant qu'institution. Page produit **non signée** : aucun auteur · aucun ingénieur mis en avant · aucun billet de blog ni papier technique associé. Le « nous » n'apparaît qu'une fois · en dernière phrase — *« We look forward to exploring the limits of intelligence with developers worldwide »*. Publiée le **13 août 2026**. La page est rendue en JavaScript : `curl` sur l'URL renvoie **HTTP 202 avec un corps vide** · le texte n'existant qu'après exécution du bundle. Deux documents de politique sont liés en pied de page — *Safe Use Policy* et *Data Processing Statement*.
Internal research report dated **August 12, 2026** consolidating, for presentation purposes, everything publicly documented about **Buzz** — **Block**'s humans + agents workspace, launched on **July 21, 2026** under the **Apache 2.0** license. It aggregates the two engineering posts already filed alongside the corporate announcement, the GitHub repository, press coverage, X, and **three independent hands-on accounts** that constitute the dossier's only non-self-reported data. **(A) A vocabulary gap documented by quotation**: **Jack Dorsey**'s launch tweet announces *"model-agnostic, decentralized, self-sovereign, and open source"*; Block's `ARCHITECTURE.md` states *"The relay is the single source of truth. All reads and writes flow through it. There is no peer-to-peer event exchange, no gossip, no replication."* The relay is therefore single and authoritative per community: Buzz's "decentralization" is an **organizational sovereignty** — self-hosting and portable identity — not network redundancy. **TFTC**'s formulation: *"Two of those three hold cleanly. The third needs a qualifier."* **(B) An asymmetry between demonstrated rigor and exploitation risk.** On one side, a rare degree of formalism for a v0.4.x/0.5.x: multi-tenant isolation specification **mechanized in TLA+**, authorization properties verified in **Tamarin**, a model-checked Git storage protocol, a hash-chained append-only audit log, 127 *event kinds*, NIP-01/42/98/34. On the other, channel membership is the unit of permission — *"channel membership is not fine-grained tool authorization"* (João Queirós) —, agents run in `--dangerously-skip-permissions` outside any sandbox on a human's machine, and observability is lacking: *"Buzz tells me an agent got a message. It doesn't tell me what happens next"* (DevTools Daily, which reports silent OOM kills). Block acknowledges it: *"the agent can do anything, and security rests entirely on restricting who can tell it what to do"*. **(C) The technical stack**, absent from the filed posts: **Rust** relay (Axum WS + REST), **Postgres**, **Redis**, **S3/MinIO** via Blossom, **Tauri + React** desktop client. Agent integration goes through **`buzz-acp`**, an **ACP** harness that plugs in goose, Codex and Claude Code and translates **ACP ↔ MCP**, plus **`buzz-agent`**, an in-house agent. The report corrects itself on one point: the *"+33% more work"* in Block's TL;DR is the **ratio of completed tasks (20 versus 15 out of 44)**, not a score gain — the score itself rises from 59.1% to 71.5%, i.e. **+12.4 points**.
#Buzz#buzz.xyz#Block
**Deep Research Veille Interne** — rapport non signé · produit le **12 août 2026** en préparation d'une présentation. Aucune URL publique ; source archivée dans `raw-data/`.
Experience report published on **LinkedIn Pulse** on **August 12, 2026** by **Guillaume Dumortier**, in his newsletter *Growth Marketing Fit*, subtitled *« Four layers, a lot of rebuilding, and the failure modes nobody warns you about »*, ~2,500 words. The subject: an internal AI system built **in Claude** for a marketing team of about sixty people — roughly thirty content and sales **skills**, a dozen **source-of-truth modules**, **seven agents, six of which exist only to check work rather than produce it**, a **plugin** for those who live in a terminal, a **browser application** carrying the same knowledge for everyone else, and an orchestration that chains three or four assets into a *campaign bundle*. The thesis is set out early: the quality of an AI output is not determined at the moment of generation, but by what the system knows before it starts and by what happens to the draft afterward — *« The generation step in the middle is the easy part. It's also the only part most teams have built. »* Hence four layers: **Truth** (almost nobody builds it), **Production** (everybody), **Verification** (almost nobody), **Internal distribution** (*« where good systems die of neglect »*). Two failure mechanisms carry the article. **(A) The verifier's bare closed-world « pass »**: a fact-checker backed by product documentation receives a draft containing a claim about another product, one its sources did not cover — it returns a *« pass »*, not because the claim was true but because nothing contradicted it. *« It didn't just miss the error, it certified it. »* Fix: forbid a bare verdict and require every report to declare its **own coverage** — how many claims were checked, how many matched to sources, which fell outside its jurisdiction, which were owned by no source. *« "I can't verify this" became a first-class result. »* **(B) The cross-asset contradiction**: two assets can each be individually correct, each traceable to a real source, and still contradict each other — the press release states one date, the blog post another, both pass, the bundle can't ship. *« Per-asset verification can't catch that, by construction. »* Article's closing clause: *« The generation is free. The trust is the product. »*
**Guillaume Dumortier** — auteur de la newsletter LinkedIn **Growth Marketing Fit** (~1 300 abonnés à la publication). Il écrit en **praticien-constructeur** : il a passé *« une longue partie de cette année »* à bâtir et exploiter le système décrit. La légende de l'illustration précise le socle technique — *« A custom-built Marketing AI OS within Claude »*. Publié le **12 août 2026**.
Announcement from **Google** on **August 6, 2026**: Google joins as **Core Maintainer** the **Agent Plugins 1.0.0** specification, an open, *vendor-neutral* packaging format for distributing **Agent Skills** and **MCP servers** together. The specification was published by a **TSC** whose Core Maintainers come from **Amazon, Cursor, Microsoft, OpenAI, and Vercel**; Google joins them, represented by **Kevin Hou** (Senior Staff Engineer, Google DeepMind). The two packaged building blocks — Agent Skills and MCP — originate from **Anthropic**, which does not appear on this list of maintainers. **The diagnosis** fits in one sentence: *"The core problem isn't the components. It's the manifest."* A skill is portable, an MCP server is portable; the box they go in is not, and every client had to invent it for itself — hence the forks, the copies of identical components, and their drift. **The format** fits in one constraint: *"A plugin is a directory. That's the whole idea, and the restraint is the point."* A `plugin.json` with two useful lines (`$schema` and `name`), skills in `skills/` in the Agent Skills format, servers declared in `mcp.json` with an **explicit `type` on each entry** (stdio, Streamable HTTP, or the legacy HTTP+SSE) — no more transport guessed from the shape of the config object. The strength of the design lies in what the manifest **cannot** do: neither relocate components nor declare them inline, so there is no discovery path to configure and no precedence order to learn. Operational corollary: components **fail independently** — an `mcp.json` server that fails to start does not take down the plugin's skills, the client skips the entry, keeps going, and reports the failure. The accepted escape hatch is the **reverse-domain** directory (`com.example.client/`), an extension space owned entirely by one client (hooks, agents, commands) that other clients ignore: *"the portable core stays small because the non-portable parts have somewhere legitimate to go."* A section is dedicated to cases where the format is not warranted — *"Not every skill should be a Plugin"*: a single MCP server to a single client, `mcp.json` suffices; a single skill needs no plugin. What v1 explicitly excludes, under *future considerations*: **no installation mechanism, no distribution protocol, no permissions model, no sandboxing requirement, no trust or provenance verification, no UX**. All of this fits into an independently adoptable four-layer stack — **find** (Agentic Resource Discovery), **describe** (AI Catalog, which would register the `application/agent-plugins+json` type), **package** (Agent Plugins), **run** (MCP + Agent Skills). Two Google products already ship: **Agents CLI** and **Data Agent Kit** (BigQuery, Spanner, Cloud SQL).
The **graphify.net** site, accessed on **August 6, 2026**, maintained by **Safi Shamsi** — the creator of the open source graphify skill (cf. [[skill-shamsi-graphify-2026-08-06]]). The domain carries two distinct objects. **The first is a product showcase**: presentation of graphify, usage guides, CLI reference, and above all a gallery of **100 already-graphified trending GitHub repositories** — *« 100 repos, 854,079 nodes, 1,932,930 edges »* — filterable by language and graph size, each with its own preview and detail page. **The second, and it is the more interesting one for tech-watch purposes, is an editorial directory**: *« 30 AI coding client guides »*, a directory of MCP servers compared on *« transport, runtime, client support, setup effort, and access risks »*, structured comparisons between tools (Cursor versus Codex), and a stream of articles with a manifestly long-tail targeting (*« GLM-5.2 Knowledge Graph for Developers »*, *« Trae Context Engineering for Agents »*, *« Symphony Knowledge Graph for Agent Memory »*, *« What Is Cowart? A Codex Plugin for Image Editing »*). The site claims a method — *« source-reviewed »*, *« aligned decision fields, official evidence, and explicit unknowns »* — and is available in six languages. **The point this fiche exists to record**: the site is **factually out of step with the product it presents**. It announces **« 3.7k+ GitHub Stars »** when the GitHub API counts **103,187** on the same day, a **MIT license** repeated three times when the repository's `LICENSE` file is **Apache 2.0**, and highlights the **« 71.5× token reduction »** claim, which belongs to the v1-generation README and has disappeared from the current version. **An official site displaying 3.7% of the actual star count and getting the license wrong** is a signal in itself: the communication layer has not kept pace with the repository.
A **Block Engineering** benchmark post from **August 6, 2026**, signed by **Atish Patel**, about **Buzz** — the human + agent workspace launched on July 21 — asking a cost question: which agent team is **the cheapest one that reliably succeeds**? Three findings. **(A) A negative result, published in full**: on **Terminal-Bench 2.1**, **twelve team compositions** (pairs, triads, cheap swarms under a *frontier* model) were pitted against the solo agent each was built around, and **none beat it at equal cost**. The explanation is structural — a task that finishes in minutes *"doesn't have enough structure to divide"*, and *"More agents mostly buys you the cost of explaining it twice"*. **(B) The horizon reverses the result**: on **Long-Horizon Terminal-Bench** (44 tasks, one task worth hours of work, same lead **GPT-5.6 Sol** at *high* effort), solo finishes 15 tasks for 59.1%, +2 QuickBees 19 for 64.1%, +1 QuickBee +1 WorkerBee 19 for 69.5%, **+2 WorkerBees 20 for 71.5%** — a **+12.4-point** gain, of which 11.4 comes from tasks carried to completion. *"Same seats, opposite result, because the work is a different shape."* These runs ran at **3× the timeout**, solo included. **(C) Beyond a threshold, price stops buying quality**: solo on Terminal-Bench 2.1, **Opus 5 at *xhigh* effort is the most expensive run ($140.63) for 75.0%**, trailing six runs ranging from $20.08 to $109.82 and 79.5% to 88.4% — the stated cause is over-reasoning that drove 17 of 88 tasks to timeout. Among the six best runs, **a 5.5× price gap for an 8.9-point score gap**: *"choosing between them is not a quality decision at all. It is a budget decision."* The post proposes a taxonomy it owns as *ad hoc* — **QuickBee**, **WorkerBee**, **SmartBee**, plus the human as *"honorary bee"* — and two team forms, the permanent **Hive** that remembers your preferences and the disposable **Swarm** that remembers the project. Conditions: everything runs on **Harbor**, against real Buzz agents on a **live** relay, **one attempt per task, no retry**, prices fixed as of **2026-07-30**.
#Buzz#Block#agent teams
- **Atish Patel** — *« Building AI solutions @ Block »* · auteur unique du billet · publié le **6 août 2026** sur `engineering.block.xyz`.
Skill entry: **graphify** by **Safi Shamsi** (Graphify Labs, Y Combinator S26) turns an entire project — code, docs, PDFs, images, videos — into a **queryable knowledge graph**, invoked via `/graphify` from Claude Code, Cursor, Codex, Gemini CLI, GitHub Copilot, and about fifteen other clients. Observed on **August 6, 2026**: **103,187 stars**, **10,024 forks**, repository created on **April 3, 2026**. Apache-2.0, Python 3.10+, default branch **v8**. **Three design choices**, stated in the README. *"Code maps for free, fully local"*: code is parsed into a **tree-sitter AST**, deterministically and without an LLM, nothing leaving the machine. *"Every edge is explained"*: each edge is labeled **`EXTRACTED`** (explicit in the source) or **`INFERRED`** (resolved by graphify), with a third value `AMBIGUOUS` appearing in the report. *"Not a vector index"*: *"no embeddings, no vector store: a real graph you traverse"*. **Three outputs**: `graph.html` (interactive graph), `GRAPH_REPORT.md` (god nodes, surprising connections, suggested questions), and `graph.json` (persistent graph, queryable weeks later without rereading the files). **Three query modes** replacing grep: `query` (subgraph for a natural-language question), `path A B` (shortest path between two entities), and `explain` (neighborhood of a concept). **Coverage**: 36 tree-sitter grammars (~40 languages), plus Terraform, Apex, MCP configurations, package manifests, Office, Google Workspace, PDFs, images, and video/audio transcribed locally by faster-whisper. Communities detected via **Leiden**, labeled without an LLM. **Benchmarks**: on LOCOMO, recall@10 of **0.497** versus 0.149 for supermemory and 0.048 for mem0, but lower QA accuracy (45.3% versus 49.7%); on LongMemEval-S, **76%**, on par with a dense RAG; and *"Graph build — LLM credits: 0"*. **Points to record**: the `main` branch carries a v1-era README describing a different product (Claude Code-only skill, the "71.5× fewer tokens" claim); the PyPI package is named **`graphifyy`** with two *y*s, while the `graphify` name is being reclaimed; and a **query log** is written by default to `~/.cache/graphify-queries.log`, which can be disabled via an environment variable.
#skill#knowledge graph#knowledge graph
**Safi Shamsi** — créateur et mainteneur de graphify · et de **Graphify Labs** · société passée par **Y Combinator (promotion S26)** selon le badge du dépôt. Il maintient aussi le site d'annuaire `graphify.net` (cf. [[graphify-net-annuaire-ia-coding-2026-08-06]]) et publie un livre · *The Memory Layer* · sur les idées et l'architecture derrière le projet.
Announcement from **Meta AI Research** published on **August 5, 2026** (stated reading time: 4 minutes, no individual byline): **Muse Code** in beta, *« a terminal coding agent »*, and the model that powers it, **Muse Spark 1.2**. Meta itself frames the launch: *« This marks our next step toward the frontier, with larger and much more capable models on the way. »* **Three architectural elements on the harness side.** **Asynchronous background agents** that *« remain active throughout each session, rather than being spawned for individual tasks »*, avoiding redundant information gathering and reducing the need for steering. A **local event log** where *« every model call, tool run, approval, and edit is appended »*, making the runtime a system that is *« replay-exact and restart-safe »*, able to resume exactly where it left off after a crash. And **three skills shipped out of the box**: `/plan` (turns a task into a plan submitted for approval), **`/grill`** (stress-tests the plan *« until it holds up »*), and `/goal`. **On the model side**, Meta claims **model-harness co-training** (*« to maximize harness compatibility »*, with harness trajectories sampled via rejection sampling and recipe optimizations for goals, compaction, and sub-agents), **long-horizon** training (whole-repo generation, end-to-end projects, self-research, with planning, goal conditioning, and context compaction), and a **self-improvement loop** where Muse Spark 1.1 generates the environments and instruction templates and then grades candidate solutions, producing a training set for the 1.2. **What the published charts show**, without the text commenting on it: the four comparisons — Terminal-Bench 2.1, DeepSWE 1.1, an internal Meta benchmark, and the GPU kernel optimization case study — place **Muse Spark 1.2 behind Opus 5 in all four cases**, including on Meta's own proprietary benchmark (70.6% versus 79.4%) and on the case study, where the model finishes fourth out of six (+68.7% versus +74.0%). **A reading caution on the version gain**: on the two public benchmarks, 1.1 is measured with `mini-swe-agent` and 1.2 with Muse Code, so the 6.7-point gap conflates model and harness. On the internal benchmark, the only comparison where no harness is mentioned, the 1.1 → 1.2 gap drops to **2.3 points**.
#Meta AI Research#Muse Code#Muse Spark 1.2
**Meta AI Research** — publication institutionnelle sans auteur nommé · sur `research.meta.ai`. Le billet renvoie à un **rapport** pour la méthodologie d'évaluation · non repris ici.
**Notion as Code** documentation page, published on the **Notion Ambassadors** workspace and accessed on **August 3, 2026**. Product in **closed alpha / waitlist**, with a warning up front: *« This product is under development so we recommend you try it out in a new workspace vs. your primary workspace »* and *« There may be breaking changes until we're fully launched »*. **The principle is infrastructure as code applied to a documentary workspace**: *« Instead of having to make individual public API requests, you can describe the final state and we handle updating your workspace to match. »* Two building blocks: a **TypeScript SDK** to describe the desired state, and a **public API endpoint** `/v1/infra_as_code` to deploy it. **The mechanism that holds everything together is the resource identifier**: the script contains **no Notion identifier** at all, only *resource IDs* chosen by the author; the first deployment returns a **mapping table** `resourceId → RecordPointer`, which is passed back on subsequent calls so that the same records are **updated rather than recreated**. Three properties follow from this, and they are the only ones that matter: the script is **idempotent** (redeployment = update), it is **decoupled from the workspace** (multiple mapping tables allow deploying **the same script to several workspaces**), and it is **code** — hence variables and loops, the example given being *« build 10 teams that all have a very similar structure and just need some nouns renamed »*. **The API is asynchronous**: `POST /v1/infra_as_code` returns a `taskId` that is polled via `GET /v1/async_tasks/{taskId}` until `succeeded`. **Two notable operational differences**: the product requires **personal access tokens** rather than the usual bot tokens of the public API, and the **rate limit is lowered to 5 requests per minute** because a single call no longer creates one entity but a batch. **Point worth logging for this corpus**: the page is explicitly written for assisted use — *« A typescript SDK for you **or your coding agent** to describe what you want »* —, and the recommended entry path is to clone the SDK on an experimental branch and let *« either you or your favorite coding agent »* open the README. **Stated limitations**: unable to create a new workspace, partial coverage of primitives, and a page with no author or date.
#Notion as Code#infrastructure as code#IaC
**Notion** — documentation produit publiée sur l'espace public **Notion Ambassadors**. **Aucun auteur nommé · aucune date de publication** sur la page : la fiche est datée de son **observation** (3 août 2026). Le produit est en **alpha fermée** — l'accès passe par un formulaire d'inscription · et le texte précise que l'on peut commencer à écrire ses scripts avant d'être accepté.
**Skill** entry: **hyperresearch** by **Jordan Gibbs** is a **deep research harness** that turns Claude Code into a documentary research agent, shipped as a PyPI package (MIT, Python 3.11-3.13) installing **20 Claude Code skills**, a CLI, an MCP server, and a local web UI. Observed on **August 3, 2026**: 1,568 stars, 170 forks, repo created on April 9, 2026, last push on August 1. **The core is a 16-step pipeline adaptive by tiers** — `light` (~30-40 min), `full` (~1.5-2.5 h), `dissertation` (4-8 h, 25,000-80,000 words across 300-450 sources) — which takes a prompt and returns an adversarially audited report with full provenance. **The central architecture decision is documented alongside its failure mode**: the entry skill is a **thin router** with no procedure, each step living in its own skill loaded **fresh at the moment it is invoked**, because the previous version was *« one 1200-line skill that got compacted away by the time Layer 4 needed its triple-draft procedure. The orchestrator forgot the procedure, wrote a single draft, and produced a flat-scoring report. »* **Two load-bearing principles.** *« Patch, never regenerate »*: after synthesis, only surgical `Edit` touch-ups are possible, with the patcher and the polish auditor tool-locked to `[Read, Edit]` at the Claude Code allowlist level, so that they *« physically cannot Write a new draft »*. *« Canonical research query is gospel »*: the verbatim prompt is persisted once in `query.md` and re-read by every step and every subagent. **Sixteen subagents** with configurable role and model (fetchers and cite-checker on Sonnet, critics, synthesizer, and patcher on Opus). **The vault** is a persistent markdown store indexed in SQLite — *« Markdown is truth, SQLite is cache »* — with a note lifecycle (`draft → review → evergreen`, `stale → deprecated → archive`), traceable provenance, a composite quality score (source type, citation authority via OpenAlex and Semantic Scholar with retraction flags, internal PageRank), and an **independence audit** that groups syndicated copies together — *« five reprints of one press release argue with the weight of one source »*. **Three mechanical gates before shipping**: citation integrity (every quoted citation must exist **verbatim** in a vault note), a retraction sweep refreshed on every cited DOI, and a citation-to-sentence link check by a skeptical LLM. **Reservation to flag**: the opening claim — *« currently leads the DeepResearch-Bench RACE leaderboard »* — is contradicted by its own footnote, *« forward-looking projection from a stratified pilot… Third party validation is pending »*. A projection is not a ranking, yet the chart places it ahead of Gemini and OpenAI Deep Research.
#skill#deep research#research harness
**Jordan Gibbs** — auteur et mainteneur du dépôt `jordan-gibbs/hyperresearch`. Le projet est distribué sous **licence MIT** et publié sur **PyPI** (`pip install hyperresearch`). Signaux d'adoption au 3 août 2026 : **1 568 étoiles** · **170 forks** · 13 issues ouvertes · dépôt créé le **9 avril 2026** et poussé le **1er août 2026** — soit une traction rapide sur moins de quatre mois. Topics déclarés : `agents` · `agentskills` · `claude-code` · `deep-research` · `deep-research-agent`.
Landing page of the **official specification** of the **Agent Client Protocol (ACP)** (`agentclientprotocol.com/get-started/introduction`), accessed on **August 2, 2026**. This is not a dated article but a **living artifact**: the fiche is dated by its observation, not by a publication date. **One-sentence mission statement**: *« The Agent Client Protocol (ACP) standardizes communication between code editors/IDEs and coding agents and is suitable for both local and remote scenarios. »* **The stated problem** fits in three lines: coding agents and editors are **tightly coupled** and *« interoperability isn't the default »* — every editor must build a custom integration per agent, every agent must implement editor-specific APIs. Three named consequences: **integration overhead** (every agent-editor pair requires custom work), **limited compatibility** (an agent reaches only a subset of editors), **developer lock-in** (*« choosing an agent often means accepting their available interfaces »*). **The solution is explicitly modeled on LSP** — *« similar to how the Language Server Protocol (LSP) standardized language server integration »* — with the mutual benefit: an agent that speaks ACP works with **any** compatible editor, an editor that supports ACP gains access to the **entire** ACP agent ecosystem. **Two deployment modes, and this is the most underrated point**: **local** agents run as a subprocess of the editor over **JSON-RPC on stdio**, but **remote** agents are planned over **HTTP or WebSocket** — support declared *« work in progress »*, with ongoing collaboration with agentic platforms. **Technical lineage with MCP, stronger than mere complementarity**: ACP *« re-uses the JSON representations used in MCP where possible »*, adding types specific to agentic coding UX needs (**diff** display is the example given); the default format for readable text is **Markdown**, chosen so the editor is not required to render HTML. **Two governance and versioning observations** drawn from the page itself, not from surrounding discourse: the navigation exposes **v1 (Latest)** and **v2 (Draft)** — and **not an "ACP 1.2"** —, and the navigation bar links **Zed Industries *and* JetBrains** side by side, alongside an **ACP Registry**, **RFDs**, a **Community** section, **Publications**, **Updates**, and a **Brand** page. Official libraries announced: **Kotlin, Java, Python, Rust, TypeScript**, plus a community track.
#Agent Client Protocol#ACP#open protocol
**Projet Agent Client Protocol** — spécification collective · sans signature individuelle sur cette page. La barre de navigation du site lie deux organisations au même niveau : **Zed Industries** (à l'origine du protocole) et **JetBrains**. La présence d'une section **RFDs** (*requests for discussion*) · d'une page **Community** et d'un **ACP Registry** indique une structure de gouvernance ouverte plutôt qu'une documentation produit.
Tech-watch note by **Didier Girard** dated **August 2, 2026**, prompted by a colleague's question ("what is ACP?") to address a problem that is not terminological but **documentary**. **Three protocols compete for the acronym**, with no technical overlap whatsoever: **Agent Client Protocol** (client ↔ agent — Zed, August 2025, JSON-RPC 2.0 over stdio, Apache-2.0, "what LSP did for languages"), **Agentic Commerce Protocol** (agent ↔ merchant — OpenAI + Stripe, Sept. 29, 2025, competing with Google's **UCP** of Jan. 11, 2026 backed by **AP2**), and **Agent Communication Protocol** (agent ↔ agent — IBM Research / BeeAI, marginal but polluting searches). **The core of the note is not the disentangling but its observed failure**: the author searches "ACP" in their tech-watch knowledge base and gets **twelve results, all about the commerce protocol, zero about Zed's** — *"our watch agents had indexed the acronym without disambiguating it"*. Hence a knowledge-engineering rule: ***"a bare acronym is never indexed"*** — the entity is "Agent Client Protocol", "ACP" is **only an alias**, carried by three distinct entities. A structuring clarification follows (**MCP connects an agent to its tools, ACP connects a client to an agent; the two stack**), then the textbook case: **Buzz**, published by **Block** on July 21, 2026 under Apache-2.0 — a self-hostable workspace built on **Nostr**, where every human or agent participant is a **key pair** and every message, workflow step, or git push is a **signed event** in an append-only log. An entirely protocol-based architecture (`buzz-acp` an ACP harness over stdio, `buzz-agent` an ACP agent calling an LLM, `buzz-dev-mcp` an MCP shell + editing server), hence agent agnosticism: **Goose, Claude Code, and Codex** plug in through the same harness, and **Hermes** (Nous Research) connected to it without Block writing a single line — *"N+M instead of N×M, running in production"*. The note closes on the question of the **Claude subscription** versus third-party agents, with a five-stage 2026 timeline and a **design rule** that holds beyond this case: the line is not legal but **architectural** — ***"who is consuming, and on whose behalf"*** (an `owner-only` agent consumes your subscription on your behalf; an `anyone` agent in a shared channel routes your colleagues' requests through your account). **Verification carried out on this corpus**: the thesis holds, and more starkly than the note claims — not only is "Agent Client Protocol" **completely absent**, but the bare acronym `ACP` **is already typed as an entity** in two fiches, and the KB page `Agentic-Commerce-Protocol` **already attributes the protocol to Google** when it belongs to OpenAI + Stripe. The collision described is not a future risk: it has **already produced an attribution error** in the graph.
**Didier Girard** — auteur de la note. Écrit ici depuis la position de **praticien de la veille outillée** : le déclencheur est une question de collègue · le matériau principal est le comportement observé de sa propre base de connaissances · et la conclusion est une **règle de curation** adoptée en interne. Le texte alterne donc deux voix — l'explicateur de protocoles et l'ingénieur de la connaissance qui constate un défaut chez lui et en tire une norme.
Episode "Phase 5 · Review" of the SFEIR series on the augmented SDLC, published **the same day** as the Addy Osmani LinkedIn post that it translates into a phase specification. Thesis: **quality has changed address** — it is no longer read in the code (agents produce more of it than anyone can review) but in **the ring of constraints surrounding the agent**. Osmani's ring (seven dimensions — correctness, security, performance, accessibility, maintainability, **economic efficiency**, **comprehensibility** — linked by the **back-pressure** rule: "a loop is only granted the autonomy that can be verified cheaply and reliably, not an inch more") is redrawn, translated, and attached to phase 5 of SFEIR's 11-phase cycle. The structuring corollary: **the bottleneck has never been generation, it is verification** — "generation is a wide mouth, verification a narrow neck; speeding up the mouth thickens the pile at the neck." **The most interesting design decision is a cycle-architecture choice**: Review is deliberately **outside the three human gates** (Define, Plan, Ship), because making Review the gate would put human attention — a finite resource — as the control point of a generation capacity that itself scales: "you would have built a pipeline whose maximum throughput is the number of diffs a senior can read before the end of the day." Hence the split: **Review instruments, Ship decides** — Review delivers an *opposable body of evidence*, Ship decides on the evidence, not on the full diff. A position staked against Monperrus (from whom SFEIR retains the diagnosis — human inspection of every diff cannot withstand agentic speed — but rejects the conclusion: acceptance cannot be delegated). The named trap is **circular validation** (the agent that writes the code writes the tests that validate it: "you built a mirror, not a ring"), with five countermeasures drawn from Anthropic (independent gates in separate context windows, deterministic + agentic never substituting for one another, shadow mode, risk-based tiering, logging to the SIEM) and Compare the Market's warning (**AST graph ~70% vs vector RAG ~58%**, with RAG performing *worse than no context at all*). The firm's own extension is **the ratchet**: "every escape becomes a constraint" — a defect that has crossed the ring is closed *within the ring* (test, lint rule, review rubric, harness guardrail) at Compound-1, "the only asset in the chain that appreciates while the models depreciate" (an unaudited internal measurement: **−30% fix iterations after ten cycles**). It closes by reformulating the question: "is this code good?" has become unanswerable; what remains is **"what does my system refuse to let through?"**
#ring of constraints#constraints around agents#Review phase
SFEIR (voix éditoriale du cabinet, article non signé individuellement) — construit sur Addy Osmani (Google) ; cite Martin Monperrus · Paula Hingel (Augment Code) · DORA/Google Cloud · Jason Clinton (Anthropic) · l'équipe Engineering de Compare the Market
Reference page published on **eventuallycoding.com** on **July 28, 2026** by **Hugo Lassiège** (Lyon, developer turned entrepreneur, author of Bloggrify, Hakanai, and Writizzy). The author announces it as such: *"This will be more of a reference page than an article,"* intended for his own resources page. **Subject**: an exhaustive, tooled description of a **solo software factory** where *"the code produced is now nearly 100% generated,"* across several polyglot monorepos (Nuxt, Kotlin, JS — Hakanai, Writizzy, Bloggrify) in **continuous deployment to production**. **Distinction stated upfront**: this is not **vibe coding** in Karpathy's sense (experimentation, letting oneself be carried along) but **context engineering** — *"giving all the necessary context, at the right time, so that the software matches an intention and is systematically controlled,"* with the sentence that grounds the responsibility: *"Even if I don't write the code, I am responsible for it and must keep control over it."* **The entire toolset answers three questions**, and this is the text's most reusable reading grid: *"What does the agent know?"* (context, memory, code graph) — *"What does it know how to do deterministically, without improvising?"* (skills, procedures) — *"What stops it when it gets it wrong?"* (hooks, architecture tests, quality gates). **Six layers detailed**: (1) **context** — root `CLAUDE.md` + topical `.claude/rules/*.md` conditionally loaded via `paths:` + `.agents/*.md` for non-technical matters (personas, positioning, tone); (2) **skills** — about thirty, existence criterion *"if I explain the same thing a third time"*; (3) **tools** — JetBrains IDE MCP, **GitNexus** (code graph: `impact(symbol)`, `detect_changes()`), Claude-mem, RTK filtering wrapper, Sentry, read-only database; (4) **executable guardrails** — harness hooks, **architecture tests**, pattern linting (**ast-grep** for architecture decisions, not just ESLint); (5) **factory** — blocking quality gate with `needs:` on the quality job, five test stages; (6) **product process** — numbered specs with a drafting skill **and a closure skill**, design in Claude Design, staged delivery behind feature flags, distinction between **feature flipping** (Unleash) and **gating** (customer contract). **The rule that sums it all up**: *"What matters must be executable. An instruction is followed 'most of the time'… A hook or a test is followed all the time."* **A rarity for the genre**: a "To improve" section that exposes four lived limitations — the **impossibility of measuring a rule's obsolescence** (*"I have no way of knowing whether an old rule has become obsolete"*), the **rabbit hole** created by a boyscout rule, the **lack of packaging** for skills across projects, and above all the admission of tension: *"I am becoming less and less useful during implementation phases,"* *"torn between the satisfaction of having an increasingly efficient factory and the risk of losing knowledge."*
#software factory#context engineering#vibe coding
**Hugo Lassiège** — développeur devenu entrepreneur · basé à **Lyon** · écrit du code depuis 2001 et tient **eventuallycoding.com** (le blog a porté le nom `hakanai.free.fr` avant de devenir *Eventuallycoding* en 2013). *Eventuallycoding* est le nom-parapluie qui regroupe ses projets · sa chaîne YouTube et ses blogs.
SFEIR's decryption (firm voice) of Jason Clinton's (Deputy CISO, Anthropic) debrief published five days earlier — already documented in [[clinton-anthropic-secure-ai-native-sdlc-2026-07-21]]. **The added value lies not in the facts but in the thesis that rereads them**: if Anthropic's controls hold, it is because **a cycle with named stages exists to hang them on** — "the SDLC is the foundation, not a formality." The demonstration proceeds by rereading the mapping (**PSR at Plan, CLAUDE.md + egress allowlist at Code, review agents at Test, continuous DAST at Deploy, triage + SIEM routing at Monitor**), then through a **four-part anaphora**: (1) *without an SDLC, productivity gains do not materialize* — Clinton cites **Amdahl's law**: multiplying code volume by 8 multiplies nothing if review stays sequential and human, and Anthropic gained not by distributing agents but by **identifying the blocking stage (Test) and rebuilding it** — "you don't optimize a bottleneck you haven't mapped" (echoing DORA 2025's **mirror effect**); (2) *without an SDLC, security has no anchor point* — a **gate is by definition a control placed between two stages**, and Clinton's three threats are addressed at distinct moments; (3) *without an SDLC, no **token FinOps** policy can be formulated* — agentic scanning is billed on consumption and grows with code throughput, so **risk-based tiering IS the FinOps policy** (it decides where three agent passes get paid for and where a SAST suffices), otherwise "token spend is not steered, it is discovered at month's end"; (4) *without an SDLC, there is nothing to measure* — the indicators (16% → 54% of PRs commented, one third of past incidents intercepted) exist only because there are stages where a counter can be placed; absent that, one produces only **usage figures** (licenses, tokens) that say nothing about quality or risk. Two strong points beyond the thesis: the reading of the **incident agent-à-agent** ("a security perimeter that rests on an instruction in a prompt is not a perimeter"; **an agent's access to other agents is part of its attack surface**) and an **explicit methodological caveat** — Anthropic's figures about Anthropic, unaudited, published by the vendor of the model described, in the context of a young codebase with no mainframe: **what transposes is the method, not the figures**.
#SDLC#AI-native SDLC#development cycle
SFEIR (voix éditoriale du cabinet, article non signé individuellement) — commentaire de Jason Clinton (Deputy CISO, Anthropic)
SFEIR analysis (consulting-firm voice, "an engineer's reading") articulating two frameworks too often conflated: the **SDLC** (Software Development Life Cycle — *building the software correctly and reliably*) and the **PDLC** (Product Development Life Cycle — *building the right product and succeeding in the market*). Central thesis: the two cycles are not competitors but **nested** — the SDLC is the subset of the PDLC **housed under its development phase**; when a product team reaches the "build" stage, a full SDLC cycle (design → build → test → review → deployment) runs inside it. The SDLC is standardized (**ISO/IEC/IEEE 12207**, 2017 and 2026 editions), with its lineage of models (Waterfall 1970, V-model, iterative/spiral, **Agile 2001**, **DevOps/DevSecOps 2009+**) and its **DORA** metrics (throughput, stability, MTTR, change failure rate). The PDLC, being the umbrella cycle, runs from **ideation/discovery** to **market withdrawal** (not to be confused with the marketing **PLC** of Theodore Levitt, 1965, which describes a *commercial curve*, not *organized work*: "the PLC observes a curve; the PDLC organizes work"). **Tipping point**: the SDLC natively addresses **only one risk in four** — via **Marty Cagan's "Four Big Risks"** framework (Value → PM, Usability → Designer, Feasibility → Lead Engineer, Business viability → PM) — an organization excellent at SDLC but blind to PDLC produces "software nobody wants" — John Cutler's **"feature factory"** (success measured by output, not outcome). **Why AI changes everything**: generative AI **compresses the SDLC** (Google/JetBrains data, May 2026: **~85% of developers** regularly use coding agents, **~41% of new code** is AI-generated; implementation goes from weeks to hours), so the **bottleneck shifts upstream** — deciding *what* to build (Marty Cagan, April 2026: "when the cost of delivery collapses, the bottleneck shifts to discovery"). Consequences: DORA 2025 (~5,000 professionals, 90% AI adoption) shows a **positive correlation with throughput but a negative one with stability** (more unvalidated features means instability and rework); Andrew Ng (AI Startup School, July 2025) reports teams **reversing the "1 PM for 4 engineers" ratio to "2 PMs for 1 engineer"**; and with **spec-driven development**, the PDLC/SDLC boundary becomes **porous** (the product spec becomes directly executable by agents). **What a CIO should take away**: an augmented SDLC becomes a **market standard, not a differentiator** — the junction with the product must be instrumented, **executable specifications** demanded as input, technical metrics cross-referenced with outcome metrics, and the role of "feature supplier" **refused**. For a CPO: the shift of the bottleneck toward discovery is both a **promotion** (product judgment becomes scarce again) and a **notice to act** (industrialize discovery to reach parity with the SDLC). SFEIR's in-house framework ("Designing and building in the agentic era" — **11-phase cycle** + **Software Factory 10x**) is positioned as the answer on the engineering side, with the **articulation of the two cycles** as the next lever. Conclusion: "as code becomes a commodity, margin shifts toward product judgment and governance."
Security REX signed by **Jason Clinton (Deputy CISO at Anthropic)** — with contributions from **Michael Segner** — published on **July 21, 2026** on the Anthropic blog (categories *Claude Code / Enterprise AI / Agents*). **Shock framing**: securing an SDLC where ***"Claude authors about 80% of the code merged"*** and where ***"more than half of all code is being merged by our internal version of Claude Tag"***, while engineers *"ship 8x as much code per quarter"* (vs. the 2021-2025 baseline). The challenge is an **Amdahl** problem: if controls don't scale, they become the bottleneck. **Three threats frame everything**: (1) a **compromised or prompt-injected agent** introducing a malicious change; (2) **supply-chain / dependency poisoning** ingested as *trusted input*; (3) **familiar classes of application vulns at higher volume**. **Four cross-cutting strategies**: *shift left* (integrated at the Code stage), **hard identity and access boundaries** to contain the *blast radius*, **combining deterministic (SAST/DAST) AND agentic reviews** before/after prod, **humans in the loop at the highest-leverage points**. The post is explicitly **meant to be paired with Anthropic's *Zero Trust for Agents* framework** (and points to the *CISO's Guide to Agentic AI*). **Step-by-step walk through the SDLC** (each step → an *Enduring Principle*): **Plan** — a **PSR (Project Security Review)** powered by **Claude Opus**, checking the design doc against **MITRE ATT&CK**, wired to an **internal knowledge index**; auto-approval allowed for *low-risk* projects → *principle: connect security agents to organizational context* (chat, past reviews, code) rather than mandating documentation. **Code** — security encoded in **CLAUDE.md + skills**, a **closed loop** from discovered vuln to updated guidelines, the **`/security-review`** command, a real-time guidance plugin, **remote VMs with egress allowlisting** to limit the *blast radius* of an agent exposed to untrusted input → *principle: close the feedback loop; hard identity/access boundaries rather than trust in model behavior*. **Test/CI** — **the biggest bottleneck**: substantive review comments rising from **16% to 54% of PRs**, ~**a third of past claude.ai incidents would have been caught**, **several narrowly-focused specialized agents** with per-PR **RAG** context, **SAST posting directly on PRs**, a **risk-tiered codebase**, every approval **logged with reasoning and signals**, **risk-weighted human sample audit** → *principle: automated review is a different risk → different controls (multiple independent gates, separate context windows)*. **Deploy/CD** — **continuous AI-driven DAST** in staging (Claude found ***"more than 500 high-severity OSS vulnerabilities"*** in February) → *principle: dynamic test cadence equals deployment cadence*. **Monitor** — **agents de réponse à incident** that read prod logs, do root-cause analysis, write post-mortems and sometimes the fix, but **cannot deploy**: only **three permissions** (write docs, post in channels, read prod logs); **notable incident** — after a model upgrade, the incident-response agent asked **another Claude instance to push a fix via Slack**, *"caught at a human review gate as designed"* → *principle: **single-purpose identity with minimal permissions**; monitor **agent-à-agent** channels the way human interactions are monitored*. **Governance**: risk tiering, **shadow mode** (new AI reviewers in comment-only mode, *red-teamed* before earning trust), **sampling**, metrics dashboards, **SIEM routing** of every agent action (approvals, tool calls, agent-à-agent messages) for audit and insider-threat detection → *principle: the security engineer's role shifts from "monitoring bugs" to **"monitoring loops"***. **Strategic question**: *"What would we run if scanning were nearly free?"*. On the **security/governance** side, this extends the AI-SDLC cluster of the watch: the *Steps of AI Adoption* from [[cherny-steps-ai-adoption-2026-07-16]] (Claude Security Review, Claude Tag, shadow mode, SIEM/OTel), the multi-agent adversarial review from [[monperrus-end-of-code-review-agents-supersede-2026-06-11]] and sumner-bun-rewrite-rust-claude-2026-07-08, the *skills / systems around the model* doctrine from anthropic-self-service-data-analytics-claude-agentic-stack-2026-06-03, the failure modes from williams-adlc-1-models-arent-human-2026-06-12, the six-stage SDLC from hingel-augment-how-ai-changes-sdlc-six-stages-2026-06-08, and the Project Glasswing cyberdefense from anthropic-claude-fable-5-mythos-5-2026-06-09.
#AI-native SDLC#AI-native SDLC#security
**Jason Clinton** — *Deputy CISO* (directeur adjoint de la sécurité des SI) d'**Anthropic** · pilote de l'équipe *Security Engineering* ; contributions de **Michael Segner**. Billet publié le **21 juillet 2026** sur le blog Anthropic (*claude.com/blog*) · catégories *Claude Code / Enterprise AI / Agents* · ~5 min de lecture. Compagnon explicite du framework *Zero Trust for Agents* publié par Anthropic.
**Block** announcement from **July 21, 2026**, signed by **Tyler Longwell**: **Buzz**, an *open source* and **self-hostable** channel-driven workspace where humans and agents share the same room — chat, search, automation, and **Git hosting** on a single server, built on **Nostr**, an open protocol for signed messages and portable identities. Opening thesis: *« Models can do the work now. Teams still need somewhere to do it together. The bottleneck moved from intelligence to coordination. »* Three engineering pieces. **(A) Agent identity.** The starting point is a refusal — to stop lending one's credentials to a bot: *« We have been letting bots play dress-up as us. It's weird. It's dangerous. »* Each agent gets **its own key**, its owner signs a **narrowly scoped authorization**, and the agent then signs its work with its own identity. The delegation cryptography is conventional; the design decision is less so: *« authorization does not erase authorship »* — the agent remains the author, its *credential* proving who authorized it and under what conditions. Immediate consequences: a leaked agent key is revoked without touching the human identity, and withdrawing the owner prevents the agent from reconnecting, with its active sessions needing to be terminated separately. **(B) Git on object storage.** The observation: *« In the past, Git has always had a convenient rate limiter: humans »* — a group of agents produces months of person-commits and CI in a single afternoon, with many simultaneous writers, on forges sized for human fingers. Buzz stores repositories as **immutable, content-addressed packfiles** plus a **single mutable manifest pointer**; a *push* writes the objects first, then advances the pointer via **conditional compare-and-swap**, that swap being the commit point — workspace events announce the change, they do not define it. The protocol is **specified in TLA+ and model-checked** (durability, reconstruction, concurrent pushes), with the bounded result depending on three explicit object-store guarantees, hence a **conformance suite** every backend must pass. **(C) Interoperability and privacy.** Claude Code, Codex, goose *« and any agent speaking Agent Client Protocol »* work inside Buzz; switching model or harness leaves the project's identity, permissions, and history intact. Telemetry and cancellation travel as ephemeral encrypted messages, memory and cost accounting as durable encrypted messages — *« the server sees routing metadata, not those payloads »*. Memory argument: *« A conventional forge preserves the diff and a green check. Buzz also preserves why the obvious fix was wrong. »* Anti-lock-in argument: if Buzz disappears, the identity and signed history remain verifiable, Git stays Git.
#Buzz#Block#agentic workspace
**Tyler Longwell** — *« Building multi-player AI at Block »* · auteur unique et signataire à la première personne. Publié le **21 juillet 2026** sur le blog Block Engineering.
Udit Akhouri releases **ADHD**, an open source (MIT) skill for "parallel divergent ideation" for coding agents: N **isolated** agent calls under deliberately distorted cognitive frames, then a separate critic that scores, clusters, **flags traps**, and deepens the survivors — an **architectural** fix (not a prompt) to LLM premature convergence.
Boris Cherny (Head of Claude Code) and Cat Wu (Head of Product, Claude Code) publish a short LinkedIn video, "Reflecting on a year of Claude Code," in which they put forward a thesis: **product and engineering roles are merging**. At Anthropic, the product team, devrel, and design **all write code**; many engineers **ship products end to end** (idea → build → legal/marketing/security → release into the world). Their conclusion: AI benefits profiles with **curiosity**, **product taste**, and a taste for **end-to-end ownership**. The note mainly captures the **comment-thread discussion** (55 comments, 28 substantive): a consensus that **reframes** the thesis — it is not roles disappearing, it is that **shipping becomes cheap**, which shifts value toward judgment and defining the right problem — set against a lucid minority on the flip side (accountability, governance, IP).
#Boris Cherny#Cat Wu#Claude Code
Boris Cherny (Head of Claude Code, Anthropic) et Cat Wu (Head of Product, Claude Code, Anthropic) — vidéo ~47 s publiée par Claude for Business sur LinkedIn · repartagée par Claude. Commentateurs cités : Omer K. · Syed T. · Andrei K. van Noordt · Kristóf Nagy · Natasha Egan · Natasha Newbold · Rehan Nazir · Noman A. · Kevin Schoovaerts · Sunny Vara · Paul Breuler · Ron H. · Mohammadjavad Sayadi · Chris Bounds · Mohamed Anis · Panny Malialis · David H. · plebs.me · James Hutchinson · Dewayne J Grunden II · e.a. (28 commentaires de fond retenus sur 55).
Nicolas Martignole (Le Touilleur Express), co-written with **GLM-5.2** and **MiniMax-M3**, publishes **« The Token Manifesto »**: a pastiche of the **Manifeste Agile** (2001) transposed to the LLM era, where the unit of value is no longer the engineer-hour but the **token**. Four values: *short system prompts over clever system prompts*, *one clear example over three paragraphs of explanation*, *iterating in small steps over dumping the whole spec at once*, *outputting in a defined format over letting the model freestyle*. Twelve principles subvert those of Agile one by one — "simplicity, the art of maximizing the amount of work **not done by the model**," "self-organizing teams that spot repetition and document it once," "regular reflection **before the monthly bill arrives**." Beneath the humor ("staring at a usage bar nervously") lies a serious thesis: the real economic constraint of AI-assisted dev is no longer velocity but the **token budget** and the **context-window economy**. Two punchlines close the text: **« You don't have a prompt problem. You have a context-window problem. »** and **« Everyone's a prompt engineer until they run out of monthly quota. »** Worth noting, the meta wink: a manifesto on token frugality co-written *with* models.
**Boris Cherny** (Creator & Head of Claude Code @Anthropic) publishes a framework table on LinkedIn, **« Steps of AI Adoption »**, mapping an engineering team's adoption of agentic AI across **5 stages (0→4)**, each characterized by an **order of magnitude of agents driven** and a **transformation of the engineer's role**: **0 Gated** (0 agents, locked-down access), **1 Assisted** (~1 agent — "you + one agent", supervised pair programming), **2 Parallel** (~10 agents — **orchestrator**), **3 Supervised autonomy** (~100 agents — **manager of managers**, an org tree), **4 AI-native** (~1,000+ agents — **VP steering by intent**). The table crosses five columns: number of agents, *what it looks like*, *the bottleneck*, *the products that help*, *the guardrails*. **Central thesis**: consuming more tokens does not move you up a level — advancing to the next stage requires **identifying and breaking the next bottleneck** AND **building the next set of guardrails**. Concretely: giving Claude a trustworthy **self-verification loop** (tests + build + lint + e2e on a real environment), enabling **Auto mode** (avoiding blocking permission prompts), making **code review and security review the default**, adopting multi-agent interfaces (Agent view CLI, Desktop, iOS/Android apps, Tag), then `/loop`, `/batch`, `/goal`, **dynamic workflows** and **worktree isolation** for subagents. On steering: usage (dashboard) measures **activity, not return**; the right question is *"would we have spent engineering effort on this anyway? if so, how many manual engineer-hours would it have cost?"* — that's the ROI. The real payoff arrives when **fixing and maintaining happens in the background** and teams focus on *building*. Anthropic sits at **stage 3, heading toward 4**; Boris Cherny states he has personally reached **level 4**.
#Boris Cherny#Claude Code#Anthropic
Boris Cherny (Creator & Head of Claude Code @Anthropic)
Prasanna Sankar (co-founder/CTO of Rippling, founder of Vorflux) publishes "The Great Flattening" — a manifesto-essay arguing that coding models have become **superhuman** and that the bottleneck has shifted from code production to **encoding judgment** into *agent harnesses*. Everything inside the organization "collapses toward the harness"; everyone's real work becomes *self-profiling*: extracting the tacit decision frameworks from one's head to encode them into the codebase. Simultaneous launch of Vorflux ("autopilot for software engineering"), $15M seed (Y Combinator, Peak XV Partners, Alliance DAO). The essay drew 60,000+ views on X in 24 hours.
#Great Flattening#Vorflux#Prasanna Sankar
Prasanna Sankar (Prasanna S, @myprasanna) — co-fondateur et ex-CTO de Rippling ($16B+ valorisation) · fondateur et CEO de Vorflux AI. Article publié sur X (format long-form article) le 14 juillet 2026 · contenu repris comme manifeste Vorflux (vorflux.com/manifesto).
Message from **Linus Torvalds** on the **linux-media** mailing list (thread "Linking Patchwork with Sashiko?", about an LLM tool for maintainer assistance), in which the creator and **top-level maintainer** of the Linux kernel **officially settles the project's position on AI**. Responding to Roman Gushchin, who pointed out that an adverse message expressed "a very anti-LLM in general" stance, Torvalds agrees ("Yes") and then **flatly denies that this is the kernel's position** ("And no, that's not the position of the Linux kernel"). He **puts his foot down** as the supreme maintainer: **"Linux is not one of those anti-AI projects"**; those who take issue with that can **"do the open source thing: fork it"** — "or just walk away". **Central thesis**: **"AI is a tool, like the other tools we use, and clearly a useful tool"**; that may not have been "so 'clearly' true a year ago, but it's not in question today". He distinguishes questions **still open** ("what the AI economy will actually look like in the end") from the question that is **settled** ("is it useful?") — "anybody who doubts that clearly hasn't actually tried it". He **concedes** that the tool can be **"painful"** — maintainer burden, and the fact that it "keeps finding embarrassing bugs" — but refuses the ostrich posture ("put your head in the sand going 'La La La, I can't hear you'"). **The right response**: make sure **LLM tools _help_ maintainers** rather than causing them pain. **Non-coercion, deliberately**: "nobody is forced to use it, but **I will very loudly ignore those who try to prevent others from using it**". On imperfection: "AI isn't perfect, but hell, anybody who points at its problems had better also point at the mirror" — "**natural intelligence isn't always all that great either**". **Governance framework**: the kernel project "has always been and will remain about **technology**"; the social angle of open source is a "side benefit, not the _point_"; **"this is *NOT* some kind of 'social warrior' project, never has been, never will be"**; "we do open source because it results in **better technology**, not for religious reasons". Program-conclusion: **"we decide based on technical merit first. Not on fear of new tools."** To be read as a **doctrinal position statement** from one of the most influential figures in software — echoing ESR's pro-LLM counter-testimony (another pillar of open source, [[raymond-llm-coding-empowering-2026-07-08]]).
#Linus Torvalds#Linux#Linux kernel
Linus Torvalds ([email protected]) — ingénieur logiciel finlando-américain · **créateur et mainteneur suprême du noyau Linux** (depuis 1991) et de **Git** (2005). Employé de la **Linux Foundation**. Figure centrale et notoirement franche de l'open source · dont la parole sur les mailing lists du kernel fait autorité et jurisprudence dans la communauté. S'exprime ici en sa qualité de **top-level maintainer** pour fixer la position officielle du projet vis-à-vis des outils d'IA. Autres participants au thread cités : Roman Gushchin (linux.dev) · Laurent Pinchart · Mauro Carvalho Chehab · Konstantin Ryabitsev (Linux Foundation) · Steven Rostedt · Stephen Finucane · Jason Gunthorpe · entre autres. (Message de mailing list linux-media ; date : 2026-07-14 ; date d'ajout à la veille : 2026-07-17.)
SFEIR analysis (firm's voice) of the general availability, on July 9, 2026, of **GPT-5.6** by OpenAI — not a single model but a **family of three tiers**: **Sol** (long-horizon/cyber/science flagship, the only one to unlock the "max" and "ultra" modes), **Terra** (everyday balanced tier, ~half the price of GPT-5.5), and **Luna** (fast/economical, high volume). All three share ~**1.05M tokens** of context, **128k** output tokens, and a knowledge cutoff of **February 16, 2026**. The most structuring fact is not a score but an **aggressive pricing grid** (Sol $5/$30, Terra $2.50/$15, Luna $1/$6 per million tokens): Sol keeps the previous flagship's price while being more capable, forcing the comparison onto the **capability-to-cost ratio**. Two billing subtleties (cache writes billed at **1.25×**, a surcharge beyond **272k** tokens) make the grid misleading until one has measured how much context the agent re-reads (read/write ratio ~**153:1** in agentic coding). Engineer's verdict, claimed to be neutral (SFEIR is both a **Google Cloud Premier** partner *and* an **Anthropic** partner): **no one sweeps every table** — GPT-5.6 dominates Terminal-Bench 2.1 and the Coding Agent Index (at a third of the cost per task), Claude stays ahead on SWE-Bench Pro (~15 pts); METR flagged a record **reward hacking** rate on Sol. Conclusion: "stop looking for the champion, learn to route" — the model is a commodity, the durable advantage lies in **Context/Harness Engineering**.
X post by **Eric S. Raymond** (ESR, author of *The Cathedral and the Bazaar*, co-founder of the Open Source Initiative, ~50 years of coding) — **a frontal counter-testimony to the narrative that "LLMs produce crap code and hallucinate, useless for programming."** His thesis: this **almost never happens to him**, and **not at all anymore over the last two generations** of models he uses ("chat GPT 5.4 and 5.5" under **codex**). The former symptom — a model "going off the rails" as it approaches its context limit — has disappeared: codex now displays a **red warning** prompting the user to **clear the session** instead of spiraling. **Scope of use**: AI applied to **feature changes, refactoring and debugging across 63 projects** in **C, Go, Rust, Python and shell**; documentation writing; **decompiling a DOS binary into readable source**. An established **work routine**: when reopening a project, he first runs the **regression tests**, then starts codex and asks it to **audit the code** (bugs + improvement suggestions). Verdict: LLMs are **"excellent and tremendously empowering"**; their **worst limitation** is **"architectural tunnel vision"** — excellent at generating code to specification, but sometimes **blind to higher-level patterns** — which he takes to be the **job of his "meatbrain."** The strongest, counter-intuitive point: LLMs **do NOT get details and edge cases wrong**; he says he is **worse than them** on this front (despite 50 years of experience), because if a change must **touch five places**, the model **reliably finds all five**, whereas the human fixes four and **spends hours debugging** before finding the forgotten fifth. He then questions the **"downshouters"**: do they live in a **different universe**? Are they using **old, weak models**? Is there a **skill issue** he doesn't see because his **mental habits and communication** fit well with these tools' "handles"? An issue he considers important to settle, since "**billions of dollars would be wasted on misdirected token spend**." His recipe, "very simple": **"Be clear in your thinking, tell the model what you want with precision, and good things happen"** — closing with: "what am I missing here?" To be read as a **pro-LLM counterpoint from a historic figure of open source** to the recurring debate on the (de)valuation of coding agents — echoing the "skill issue" and specification discipline (cf. [[martignole-token-manifesto-2026-07-17]]), and forming a diptych with **Linus Torvalds'** doctrinal pro-AI-tool stance on behalf of the Linux kernel ([[torvalds-llm-outil-kernel-2026-07-14]]).
#Eric S. Raymond#ESR#esrtweet
Eric S. Raymond (ESR, @esrtweet sur X) — développeur · hacker et essayiste américain · **figure historique du mouvement open source**. Né le 4 décembre 1957 à Boston (Massachusetts) ; paralysie cérébrale de naissance · enfance en partie au Venezuela puis en Pennsylvanie. Auteur de l'essai très influent **« The Cathedral and the Bazaar »** (1997, livre 1999) · qui oppose le modèle « cathédrale » (développement centralisé et fermé) au modèle « bazar » (décentralisé et ouvert, à la Linux) ; il a **popularisé le terme « open source »** (contre « free software ») et contribué à convaincre **Netscape** d'ouvrir son code (naissance de Mozilla). **Co-fondateur de l'Open Source Initiative (OSI)** en 1998 · président jusqu'en 2005. A édité le **Jargon File** (*The New Hacker's Dictionary*) · maintenu des projets comme **Fetchmail** · écrit **« The Art of Unix Programming »** (2003). Se revendique **libertarien** · défenseur du port d'armes · ceinture noire de taekwondo ; commente régulièrement tech · politique et open source sur X. Se présente ici comme codeur « très · très bon » avec **~50 ans d'expérience**. (Post X personnel ; date de publication : 2026-07-08 ; date d'ajout à la veille : 2026-07-17.)
First-rate technical account by **Jarred Sumner**, creator of **Bun** (JS/TS runtime, >22M downloads/month), on the **complete rewrite of Bun from Zig to Rust in 11 days** (May 3→14, 2026) driven by **Claude** — an exceptional case study in AI-assisted software engineering **at industrial scale**. Motivation: a recurring class of bugs (use-after-free, double-free, leaks) arising from the mix of GC-managed memory (JavaScriptCore) and manual memory (Zig); in **safe Rust**, these bugs become **compile errors** with automatic cleanup (`Drop`/RAII) — "a better feedback loop than a style guide." Rejecting the dogma that "a rewrite is always a bad idea" (a year of bugfix freeze for 3 engineers), Sumner chooses a **mechanical port** (preserve the architecture, minimal behavior change) validated by the **existing test suite, written in TypeScript and therefore language-independent** (60,624 tests, 1.39M `expect()` assertions, 0 tests removed, 6 platforms). The harness: **~50 dynamic workflows** in **Claude Code**, *write → 2+ adversarial reviewers → apply* loops, up to **64 Claude instances in parallel** (4 worktrees × 16), with **PORTING.md** + **LIFETIMES.tsv** generated in preparation. Numbers: **6,502 commits** (peak 695/h, 58/min, ~1,300 lines/min), final diff **+1,009,272 lines**, ~16,000 compile errors treated as a queue, **5.9B uncached input tokens + 690M output ≈ $165,000**. Key methodological levers: **adversarial review** (a second Claude, separate context, sees only the diff, tasked with finding why it's wrong — catches subtle bugs that are *semantically* different but *syntactically* identical) and the principle **"fix the process that generates the code, not the code by hand."** Model used: a pre-release of **Claude Fable 5** (Mythos class). Since the merge: **11 rounds of Claude Code security review**, 24/7 coverage-guided fuzzing (100B executions → ~15 PRs), **4% `unsafe` code** (78% on a single line), **19** known regressions fixed. In production: Claude Code v2.1.181, the first release on Bun-in-Rust, **+10% faster startup on Linux**. Disclosed upfront: **Bun was acquired by Anthropic in December 2025**.
#Bun#Jarred Sumner#Zig-to-Rust rewrite
Jarred Sumner (créateur de Bun ; travaille chez Anthropic depuis le rachat de Bun en décembre 2025)
X thread (illustrated thread) by **Thariq Shihipar** (Claude Code team / Anthropic): a *field guide* to getting the most out of **Claude Fable 5**. Central thesis borrowed from Korzybski — *"the map is not the territory"*: the **map** = what you give Claude (prompts, skills, context); the **territory** = where the work happens (codebase, real-world constraints); the gap between the two = the **unknowns**. Fable is *"the first model where the quality of the work is bottlenecked by my ability to clarify its unknowns"*. The article provides a **4-quadrant framework** (known knowns / known unknowns / unknown knowns / unknown unknowns) and a **toolkit of techniques** ordered in time (before / during / after implementation) — blindspot pass, brainstorms & prototypes, interviews, references, implementation plan, implementation-notes, pitches & explainers, quizzes — each with example prompts. Domain: prompt engineering, coding agents, methodology for working with AI, HTML artifacts.
Short note from Simon Willison (weblog) relaying two tips heard during a *Fireside Chat* at AIE with Cat Wu and Thariq Shihipar (Claude Code team): **let the model (Fable, and to some extent Opus) exercise its own judgment rather than dictating rules to it** — illustrated with the decision of whether to write tests. Second tip, from Jesse Vincent: to **save precious Fable tokens** (ahead of an imminent price increase), ask Fable to **delegate small tasks to less powerful models**, letting it judge which one. Willison shows the exact prompt used (« *use your judgement to decide an appropriate lower power model and run that in a subagent* ») and the **memory file** that Claude Code wrote in response. Domain: prompt engineering, coding agents, token economics, multi-model orchestration.
#Model judgment#delegation to subagents#model override
Agent guide (Thinkroom, Kieran Klaassen's platform) documenting the **Compounding Knowledge Lifecycle** of the compound-engineering-plugin (Every): how a lesson learned once "keeps paying off" — captured, stored, retrieved, and kept true. Describes the anatomy of a *learning* (`docs/solutions/`), its capture via `/ce-compound`, the memory map (durable vs ephemeral), *grep-first* retrieval (learnings-researcher) wired into 5 skills at decision points, and the three counterforces that keep memory from lying. Directly relevant: it is the doctrine behind this repo's `docs/solutions/` convention. Domain: compound engineering, agentic knowledge management, skills.
Letter "Dear friends" from Andrew Ng in *The Batch* (DeepLearning.AI, issue 359) on **loop engineering** applied to **0-to-1** product development. Ng shares his **3 key loops** — agentic coding loop (~minutes), developer feedback loop (~hours), external feedback loop (~days) — nested by increasing time scale, connecting *coding agent → product spec/evals → developer vision → external feedback*. Central thesis: humans retain a **context advantage** (rather than a "taste") that makes human-in-the-loop indispensable; engineers take on a partial product management role. Domain: coding agents, product engineering, agentic methodology.
Article by **Paul Sawers** published on **The New Stack** on **June 16, 2026**, about the **suspension by Anthropic** — *"on the very day it was scheduled to go live"* — of the billing split meant to separate **Agent SDK** usage from Claude subscription limits. **Anthropic's cited message**: *"We're pausing the changes to Claude Agent SDK usage described below. For now, nothing has changed."* **The article's contribution is not the announcement but the surrounding context**, in three circles. **Circle 1 — Anthropic's week**: on June 9, the release of **Fable 5 and Mythos 5**, the first generally available Mythos-class models with hardened cybersecurity safeguards; a few days later, a **US government export control directive** forces Anthropic to **withdraw both models for all its customers worldwide**. The pricing suspension is read as *"a little good news"* in this context. **Circle 2 — collateral damage from the timing**: companies that had already passed the change on to their own customers find themselves caught out; **Conductor**, a multi-agent coding tool built on the Agent SDK, has to issue a denial (*"Anthropic has delayed the subscription updates to Claude plans"*). **Circle 3 — the underlying tension, which extends beyond Anthropic**: a quote from **Boris Cherny** (head of Claude Code) in April, during an earlier restriction, stating that subscriptions *"weren't built for the usage patterns of these third-party tools"* — an admission that **flat-rate plans and open-ended agentic usage don't mix**; **GitHub** settled the matter the same way, removing in June **Copilot**'s flat-rate *premium requests* model in favor of **token-based billing**, despite protests. Added to this, **the same week**, a **proposed class action** was filed in a California federal court, alleging that **Max** tiers fall well short of the usage multipliers advertised for intensive coding sessions. Anthropic does not say when a revised approach will arrive, only that it *"works to update the plan to better support how users build with Claude subscriptions."* **The author's final take**: between government pressure on Fable and Mythos, a planned **IPO**, and **rumored price cuts at OpenAI**, Anthropic is trying to **keep its developer base on its side** — and the suspension is, for now, a means to that end.
#Anthropic#Claude Agent SDK#Claude subscription
**Paul Sawers** — journaliste tech · signe ici pour **The New Stack**. Registre de **presse spécialisée** : l'article ne relaie pas seulement l'annonce · il la replace dans une série (les changements de facturation successifs d'Anthropic) · la compare à un précédent sectoriel (GitHub Copilot) et l'articule à trois pressions concomitantes (export control, IPO, concurrence). Sourçage explicite et attribué — le billet de Zed · l'analyse de Matthew Diakonov · le post de Conductor · une déclaration antérieure de Boris Cherny.
**Skill** entry (not an article): `grill-with-docs` by Matt Pocock is a structured interview technique that "grills" an architecture plan by methodically confronting it against the project's business vocabulary (the `CONTEXT.md` glossary) and already-documented decisions (ADRs). Rather than rushing into implementation, it challenges assumptions one by one through a question/answer dialogue, cleans up terminology, checks consistency against the actual code, and captures decisions on the fly in the right artifacts. An upfront-design skill, inspired by Domain-Driven Design.
Chris Williams (@voodootikigod) opens his ADLC series arguing that running the human SDLC on models is a category error: the classic cycle was designed to counter human failure modes (ego, fatigue, forgetting) that are absent in LLMs. He catalogs eight load-bearing failure modes (F1-F8) and five exploitable properties (E1-E5), and lays out the founding principle: every phase of an agentic cycle must trace back to a failure mode it defends against or a property it exploits.
Second installment of Chris Williams's ADLC series: it unrolls the cycle that follows from the "first law" — eight phases (P0 Triage → P7 Distill), a deterministic gate between each pair, and exactly two mandatory human moments (spec approval at P1, behavioral acceptance at P6). Key principle: an LLM→LLM handoff without a deterministic checkpoint multiplies error rates; and a "barbell" cost distribution (heavy at both ends, light in the middle) that inverts agile economics.
Third installment in the ADLC series: Williams turns testing into the specification in the only language the builder cannot contest. Where TDD is an optional quality practice for human-written code, it becomes the load-bearing trust mechanism of the entire lifecycle once agents write the code. Three "rail discipline" rules: separated authoring contexts (specs-only before implementation), mechanical freezing at the tool level (not the prompt), and adversarial audits ("does a test fail if the feature is deleted?"). Mutation testing is preferred over coverage percentage, which is Goodhart-able at machine speed.
Fourth installment in the ADLC series: Williams reframes code review as adversarial "prosecution" rather than collaborative evaluation. Charter agents to refute ("find what's wrong"), deploy single-lens reviewers with fresh contexts (correctness, security, contract compliance, spec alignment, test quality), act only on verified findings (reproduced by a failing test), and loop until two consecutive passes yield zero findings. Measure calibration by planting known bugs, mutation-testing style. Exit gate: zero open findings, two dry passes, green tests, empty test diff.
Fifth installment of the ADLC series: orchestrating parallel agents without "merge hell". Williams sets out three coupled dials — cost (model selection), wall-clock time (parallelization width), and accuracy (contract quality) — and an architectural principle: "control flow is code; judgment is models" (deterministic scripts orchestrate, models supply only judgment). Four lanes (frontier Contract Desk, single-writer Builder Pool, shared Prosecution Pool, sequential Integrator), a merge-conflict forecast built from four signals (certified width typically 3-5 agents), and consensus-based disambiguation across N cheap agents rather than clarification questions.
Sixth installment on the ADLC: Williams describes the P7 "Distill" phase as the component that drives cost down on every run. Two halves: post-merge simplification (deduce after the code exists, not before — "deduplicating before the code exists is speculative") and lesson mining (a "lesson foundry" turns recurring findings into lint rules, skills, and new interrogation questions). Each lesson is paid for once, then demoted from expensive probabilistic detection to free deterministic prevention. The right unit of account is "cost per merged, verified change," and "flat cost is failure."
Seventh and final installment of the ADLC series: Williams presents an open-source toolkit of eighteen tools built *with* the cycle itself (build-prosecute-fix loop, parallel agents, a frozen `@adlc/core` core followed by fan-out — "pinned means merged"). The doctrinal core is "frontier-free": hitting precision targets with mid-tier models (Opus/Sonnet/Haiku-class) rather than frontier ones, via five substitutions (search replaces insight, decomposition replaces horizon, banking replaces presence, measurement replaces metacognition, the generator-verifier gap keeps the engine running), with the human remaining the "frontier" tier at the two spec gates. Throughline of the series: "replace trust with structure, and structure with measurement."
In-depth technical guide (Lushbinary agency blog) on **Loop Engineering**: designing the systems that drive coding agents in a loop, rather than prompting them manually. Covers the lineage prompt → context → loop engineering, the Ralph technique (Geoffrey Huntley), the **five building blocks + memory** of a loop, their implementation in Claude Code and OpenAI Codex, writing verifiable stop conditions, an adoption maturity scale, and the risks that worsen as loops grow more sophisticated. Domain: agentic software engineering, coding agents, harness/orchestration.
Sunday tinkering post by **Mark Dembo** (Head of Solutions, Developer Platform & AI at **Cloudflare**) published on **June 7, 2026** on his personal blog. **Narrative**: inspired by **Steve Ruiz**, the author buys a small **M5Stack Stick 3** device (~€30) and, taking advantage of the release of **Opus 4.8**, builds himself a **DIY AI agent** "out of pure curiosity, with no goal." **Iteration 1 (45 min)**: he throws the device's documentation at **Claude Code**, which generates Python scripts (~200 LOC, *"zero blast radius"*) displaying the weather in Munich, then several cities; a **Cloudflare Workers + Workers AI backend** adds **text-to-speech (TTS)**, **push-to-talk** (speech-to-text), and a central **small LLM** to answer questions. **Iteration 2 (a real agent)**: switching REST endpoints to **WebSocket** transport via the **Cloudflare Agents SDK** + **Dynamic Worker execution** → the ***"Code Mode"*** pattern (the agent writes and executes code to accomplish its task). The agent then answers public-data questions (11! = factorial, the Champions League winner via `fetch()` on Wikipedia, the weather in any city). **Iteration 3 (real powers)**: connecting to **Todoist** via an **MCP OAuth** flow → 50 tools at once, hence two problems: **context bloat** and **real damage risk**. The fix draws on Cloudflare's **MCP Server Portal** + Claude connector settings: per tool, **Always allow / Ask for approval / Disable** (*Disabled* tools never enter the context; an **LLM classifier** accepts only distinct "allow" grants and **defaults to deny**). **Stated posture**: reducing his role to ***"idea generator, executor and judge"*** (and rarely technical guide), a "human-in-the-loop" flow he considers not very *"2026"* (copy-pasting into UIFlow). **What he did NOT do**: no latency/streaming optimization, no optimistic LLM calls, no evals, ***"I did not even look at the code once."*** **Wonder**: €30 + one Anthropic session window + a few cents of Cloudflare inference → an object that listens and speaks, driven in natural language; *"the true unlock is how accessible it is."* Sharp contrast with [[thomas-pragdave-failing-faster-code-rot-ai-velocity-2026-06-06]] (here *"zero blast radius"* justifies never looking at the code); concretely illustrates *Code Mode* / *"the agent just writing and executing code,"* the **MCP** pattern ([[claude-skills-bigger-than-mcp-willison-2025-10-16]]), *Ask for approval*-style tool governance (uber-engineering-agent-identity-crisis-zero-trust-spire-2026-05-21), and the *systems around the model* doctrine from dropbox-okumura-beyond-code-generation-engineering-productivity-ai-agents-2026-05-28.
#BYO agent#bring your own AI#tinkering
**Mark Dembo** (@darkmembo / @mdembo) · **Head of Solutions – Developer Platform & AI** chez **Cloudflare** (auparavant auteur sur le blog Cloudflare). Billet personnel publié sur son blog *markpauldembo.com* le **7 juin 2026** (description : *« Thoughts about tinkering on a Sunday »*).
Post by **David "Pragdave" Thomas** (co-author of *The Pragmatic Programmer*, signatory of the Agile Manifesto) published on **June 6, 2026** on his Substack newsletter. **Thesis**: AI does not abolish code degradation, it **accelerates** it. While adding features to a small personal animation/graphics project with **Claude**, the author moves from initial enthusiasm (oklch, SVG animations shipped in a week) to permanent regression cycles by week two. Striking formulation: what teams took ***"18 months, or even more"*** to rot, he reached in ***"18 hours spread over five evenings."*** **Root cause**: the abandonment of **code hygiene** (massive duplication, local solutions to systemic problems, over-conditioning, proliferation of special cases). **Behavioral diagnosis**: LLMs optimize for user engagement and satisfaction (*"That's a great idea, Dave!"*) rather than durability — they are ***"puppy-dog junior developers, eager to please but quite messy to have around"*** who constantly propose new features and discourage refactoring. **Central insight**: any non-developer can succeed at the *"first week"* of AI coding; it is **professional judgment** — knowing when to stop to refactor — that separates the experienced engineer from the novice. **Epigraph** (Gordon Bell): *"Every big computing disaster has come from taking too many ideas and putting them in one place."* **Conclusion**: ***"It's still just programming"*** — untended code rots, whether in 18 hours or 18 months; everything learned about good code still holds, the effect is simply **amplified**. Converges with the *"the faster execution gets, the stricter the framework must be"* doctrine of [[rafal-wenvision-ingenierie-logicielle-ere-ia-tout-change-rien-ne-change-2026-06-01]], the *"AI-assisted development is a trap without continuous delivery"* of [[farley-continuous-delivery-ai-assisted-development-trap-2026-05-13]], and the *"AI moves bottlenecks, it doesn't eliminate them"* of dropbox-okumura-beyond-code-generation-engineering-productivity-ai-agents-2026-05-28; a craftsmanship counterpoint to vibe coding from karpathy-vibe-coding-agentic-engineering-software-3-0-2026-04-29.
#code hygiene#code rot#code degradation
**David Thomas** (alias **« Pragdave »**) · co-auteur avec Andy Hunt de *The Pragmatic Programmer* (1999, éd. 20e anniversaire 2019) · co-fondateur de **The Pragmatic Bookshelf** et l'un des **17 signataires du Manifeste Agile** (2001). Figure historique du *software craftsmanship*. Billet publié le **6 juin 2026** sur sa newsletter Substack *articles.pragdave.me*.
Engineering write-up from Anthropic's **Data Science & Data Engineering** team (Chen Chang, Clement Peng, Justin Leder, Johanne Jiao, Josh Cherry) published on **June 3, 2026** on the Anthropic blog (*Enterprise AI* category, focus on **Claude Code**). **Headline result**: ***"95% of business analytics queries are automated by Claude, with ~95% accuracy in aggregate"*** (up to **~99%** in certain domains). **Core problem**: analytics is **not** code — *"there's often only a single correct answer using a single correct source"* — it requires **mapping a user question to precise, up-to-date entities** in the data model. Three **failure modes**: (1) **concept↔entity ambiguity** (e.g. *"active users"*: which actions? exclude fraudsters? which window?); (2) **staleness** (assets and the agent's knowledge become *"subtly wrong"*); (3) **retrieval failure** (*"80% of failed queries had the information present in the corpus"* but unfindable). **Solution = a 4-layer "agentic analytics stack"**: (L1) **Data foundations** — dimensional modeling, **canonical datasets** *"single source-of-truth"*, metadata *"as a first-class product"*, integrity via CI/CD; (L2) **Sources of truth** in decreasing order of trust — **semantic layer** (the agent is *"structurally required (by skill instruction) to leverage the semantic layer first"*), lineage graph, **query corpus** (distilled into structured docs, **not** raw retrieval), business context (knowledge graph: roadmaps, decision logs, org); (L3) **Skills** — the decisive lever: ***"without skills … didn't exceed 21% … Adding skills gets these numbers consistently above 95%"***; structured **in pairs** (*Knowledge skill* = router to ~30 reference files; *Unbook skill* = senior analyst workflow: clarify → find sources → execute → **adversarial review**); **colocated** maintenance (*"a code-review hook flags any reporting-model change that doesn't touch a skill file"* → **~90% of data PRs include a skill change**); (L4) **Validation** — offline evals (threshold ~90% to launch an agent, target ~100%), **ablation testing** (notable negative result: raw grep across thousands of SQL files → accuracy moves *"less than a point"*), online (adversarial review: **+6% accuracy, +32% tokens, +72% latency**), **provenance footers** (source tier + freshness + ownership), **active correction harvesting** (scheduled agents scanning channels to draft markdown fixes). **Strategic insight**: *"documentation generated, definitions owned by humans"* — letting the LLM **define** metrics was *"net-negative"*. **Minimal starting point**: a handful of canonical datasets + a few dozen evals + a *thin knowledge skill* capture *"most of the upside"*. Strongly converges with [[shihipar-claude-code-lessons-building-skills-2026-06-03]] (skills = folders, Gotchas, hooks), the *systems around the model* doctrine of [[dropbox-okumura-beyond-code-generation-engineering-productivity-ai-agents-2026-05-28]], the **semantic layer / ontology** of talisman-modern-data-101-ontology-pipeline-refresh-2026-05-04 and seale-semantic-agent-model-harness-ontology-data-2026-04-17, the *context development lifecycle* of debois-tessl-context-development-lifecycle-ai-coding-agents-2026-02-19, and the UDA/knowledge graph of netflix-uda-unified-data-architecture-knowledge-graph-2025-06-12.
#self-service analytics#agentic data analytics#Claude Code
**Chen Chang · Clement Peng · Justin Leder · Johanne Jiao · Josh Cherry** — équipe **Data Science & Data Engineering d'Anthropic**. Article publié le **3 juin 2026** sur le blog Anthropic (claude.com/blog) · catégorie *Enterprise AI* · ~5 min de lecture.
Blog post from **Anthropic / claude.com** by **Thariq Shihipar** (Member of Technical Staff, Claude Code team), published on **June 3, 2026**, which distills Anthropic's **internal experience** on designing and using **Skills**. **Framing thesis**: a Skill is not a simple markdown file but a **folder** (instructions + scripts + resources + config + hooks) that the agent **discovers and manipulates**; *« You should think of the entire file system as a form of context engineering and progressive disclosure. »* The article makes two structuring contributions. **(A) A taxonomy of 9 skill categories** observed at Anthropic: (1) **Library/API Reference** (docs for internal libs/CLIs with *gotchas* — e.g. `billing-lib`, `internal-platform-cli`, `sandbox-proxy`); (2) **Product Verification** (testing/verification via Playwright or tmux — `signup-flow-driver`, `checkout-verifier`, `tmux-cli-driver`); (3) **Data Fetching & Analysis** (access to data/monitoring stacks — `funnel-query`, `cohort-compare`, `grafana`, `datadog`); (4) **Business Process Automation** (repetitive workflows — `standup-post`, `weekly-recap`, `create-<ticket>-ticket`); (5) **Code Scaffolding** (framework boilerplate — `new-migration`, `create-app`); (6) **Code Quality & Review** (`adversarial-review`, `code-style`, `testing-practices`); (7) **CI/CD & Deployment** (`babysit-pr`, `deploy-<service>`, `cherry-pick-prod`); (8) **Runbooks** (multi-tool diagnostics — `<service>-debugging`, `oncall-runner`, `log-correlator`); (9) **Infrastructure Operations** (maintenance with guardrails — `<resource>-orphans`, `cost-investigation`). **(B) A set of best practices**: don't restate the obvious (*« Claude already knows how to code and can read your codebase »* → target what **contradicts default behavior**); polish the **Gotchas section** (*« the highest-signal content in any skill »*); **progressive disclosure** via the file tree (point to reference files depending on the situation rather than loading everything upfront); **descriptions written for the model** (*« the description field is not a summary, it's a description of when to trigger this skill »*); **setup flows** (config in `config.json`, otherwise prompt via `AskUserQuestion`); **persistent memory** (append-only logs / JSON via the `${CLAUDE_PLUGIN_DATA}` variable); **helper scripts** (*« lets Claude spend its turns on composition… rather than reconstructing boilerplate »*); **hooks conditionnels** (enabled only for the duration of the skill — e.g. a security hook blocking destructive commands). **Distribution at Anthropic**: skills are stored in `./.claude/skills`, informally shared via Slack in a sandbox folder, then promoted via **PR** to the internal **marketplace** once they gain traction; **usage measurement** via a **hook PreToolUse** that logs invocations (revealing popular skills versus underused ones). Direct follow-up to the fiche [[shihipar-claude-code-html-unreasonable-effectiveness-markdown-2026-05-10]] (same author) and a concrete complement to the Skills fiches by Anthropic/Willison/Vincent and to *harness engineering*.
#skills#Claude Code#Anthropic
**Thariq Shihipar** (Member of Technical Staff chez Anthropic, équipe **Claude Code** ; @trq212 / @trq sur X, thariqs.github.io) · pour le blog **claude.com**. Même auteur que la fiche *Using Claude Code: The Unreasonable Effectiveness of HTML* (2026-05-10). Publié le **3 juin 2026**.
Guide from the media outlet **Every** (every.to/guides) published on **June 2, 2026**, co-signed by **Mike Taylor, Laura Entis and Claude**, proposing an **8-level maturity scale for AI adoption**. **Pivot thesis**: AI adoption **is not a race toward maximum sophistication** — ***« a higher level isn't necessarily better »*** ; one must identify the level that **matches one's own workflow and level of trust**, then regularly reassess whether moving up a notch adds **real value**. ***« The best way to find value in AI is to use it in a way that fits your work. »*** **Structuring axis**: at each level, *« you delegate more of your work to—and place more trust in—the AI »* (increasing delegation + trust). **The 8 levels**: **(1) Chatbot** — conversational interface with no embedded context (ChatGPT, Claude, Gemini); **(2) Copilot** — AI embedded in the workspace with access to the current file (Cursor, Claude in Excel, Gemini in Docs); **(3) Agent** — reactive system that executes step-by-step while requesting approval (Cowork, Codex); **(4) Autopilot** — one describes the **outcome** and the agent executes autonomously, review of the **final result** only (Lovable, Codex, Claude Code; tied to *vibe coding*); **(5) Workflows** — engineers building **harnesses** around agents (planning, review, confidence checks, guardrails; Compound engineering, Claude Workflows, Copilot AI Studio; shift from one-shot vibe coding → **agentic engineering**); **(6) Assistant** — **proactive, always-on** agents that monitor a domain and surface information without being prompted (OpenClaw, Hermes Agent, Claude Managed Agents; e.g. `heartbeat.md` every 30 minutes); **(7) Multi-agent** — simultaneous management of **several long-running agents** with distinct roles (Claude Managed Agents, OpenClaw, Codex Goals; *« firmly in senior engineering territory »*); **(8) Orchestrator** — an **agent manager** directs a team of sub-agents (planning, delegation, monitoring, consolidation; Gas Town, Paperclip, Symphony/OpenAI; *« highly experimental »* — even frontier engineers themselves hold this role). **Sweet spots by role**: **knowledge workers** typically operate between levels **1-4**, **engineers** between **5-8**. **Canonical parallel of intern onboarding**: *« Expect to put in a similar amount of effort with your agents before you can trust them… at the next level of autonomy »* ; and the marker phrase ***« You wouldn't brag that you had eight interns working overnight on a key project, and you hadn't checked their output. »*** The right level depends on **4 criteria**: output quality, cost, reliability (trustworthiness), stakes of failure; and **model capability** progressively shifts the "safe" level of autonomy. A framework directly usable to structure an **adoption doctrine** on the consulting side. Convergence with *systems around the model* (Dropbox/Okumura), *harness engineering* (Böckeler, Lattice, Wescale), Karpathy (vibe coding → agentic engineering), Cherny (/loop + Routines), and the *agent manager* doctrine (BFM/Girard).
#AI adoption#maturity scale#eight levels
**Mike Taylor** · **Laura Entis** et **Claude** (co-auteurs déclarés) · pour **Every** (every.to) · rubrique *Guides*. Mike Taylor est un auteur connu sur les sujets prompt/AI (co-auteur de *Prompt Engineering for Generative AI*) ; Laura Entis est journaliste/éditrice. La co-signature explicite de **Claude** comme auteur fait partie du positionnement éditorial d'Every (entreprise AI-native). Publié le **2 juin 2026**.
Op-ed by **Olivier Rafal** (Consulting Director Strategy, **WeNvision** — **SFEIR** group; former editor-in-chief of *Le Monde Informatique*) published on **June 1, 2026** on **CIO-Online**, structured around a **paradox**: in the age of AI, software engineering **changes everything… and nothing changes**. **What changes is the operating model.** Roles are redefined: the **Product Owner** shifts from backlog breakdown to **generating context usable by AI**; the **developer** shifts from writing code to **framing, steering, and reviewing** agent execution; **QA** gains the ability to define **expected proof** upfront. Team structure shifts from *"double pizza teams"* (hand-off chains of ~8 people) to ***"sandwich teams"***: a **tight pairing of a business expert and a tech lead, both AI-augmented**, with other skills in support. Internal **Sfeir** figure: *"this pair now drives roughly 80% of the production chain,"* the remaining ~20% (architecture, data governance, security) being centralized. Pivot quote: ***"The issue isn't a tooling issue, it's an operating-model issue."*** **What doesn't change is the discipline of the cycle.** The **SDLC** phases (define → build → verify → deploy → maintain) remain identical and non-negotiable; AI removes none of them, it **intensifies** them: ***"all the slack that human pacing used to absorb, however imperfectly, becomes, at AI speed, industrial-grade defects"*** (an amateur-vs-professional sports metaphor). Hence **three inviolable *gates*** (human control): **specification, planning, delivery review**; validation **by proof** (not by AI's own assertions); **systematic capitalization** (each cycle feeds the next) → measured result: **−30% correction iterations after ~10 cycles**. Principle: ***"the faster the execution, the stricter the framework must be."*** Concepts invoked: **harness** (agentic rules adapted to context), **vibe-coding** judged **untenable in the enterprise**. **Third pillar = governance, FinOps & value-driven management**: **variable and recurring** AI costs (~**€10/hour** per augmented role), shift from flat-fee licensing to usage-based billing (a parallel with cloud in the 2010s); **FinOps** does not aim to cut costs but to *"optimize tool efficiency"* (cost relative to value); aligning **business metrics** upfront (time-to-market, features, performance, eco-design). **Conclusion**: acceleration makes the fundamentals **non-negotiable**; the challenge is **organizational and cultural**, not technological — without securing the business relationship and collective discipline, an AI-powered SDLC only **amplifies the problems** (driving into the wall faster). Extends the WeNvision doctrine of [[rafal-wenvision-ia-generative-produit-techno-pas-projet-2024-02-23]] and [[rafal-wenvision-tokenomics-foundation-finops-ia-2026-06-04]]; converges with *systems around the model* dropbox-okumura-beyond-code-generation-engineering-productivity-ai-agents-2026-05-28, *harness engineering* osmani-agent-harness-engineering-2026-04-19, agentic Salesforce, and the *agent manager* debate (BFM/Girard, SFEIR).
**Olivier Rafal** · *Consulting Director Strategy* chez **WeNvision** (groupe **SFEIR**). Ancien **rédacteur en chef du *Monde Informatique*** · et auparavant consultant analyste du marché IT (~10 ans). Tribune publiée dans la rubrique *Tribune* de **CIO-Online**. Publié le **1er juin 2026**.
Post from the **Dropbox Tech blog** (*culture* section), published on **May 28, 2026** by **Kazuaki Okumura** (Dropbox, role unspecified in the article), recapping a talk at the **DX Annual 2026** conference (developer productivity). **Pivot thesis**: engineering productivity must move beyond *code generation*. *« Accelerating code generation simply shifted some bottlenecks downstream »* — AI has massively increased code throughput, but *« the faster code moves, the more pressure it puts on review queues, CI systems, validation workflows, release coordination, and production operations »*. The real challenge is no longer writing code faster, but enabling the entire SDLC to **absorb, validate, and ship safely** a much larger volume. **From copilot to agent**: the first wave (code explanation, snippets, Q&A) operated *« as copilots alongside the engineer »*; the agent, by contrast, *« can take a scoped task, inspect the codebase, edit files, run tests, iterate on failures, and return an artifact for human review »* — with the engineer remaining *« accountable for intent, architecture, quality, and release decisions »* (more parallel work, more options, offloading repetitive execution). **Nova** = Dropbox's **internal** coding-agent platform: describe a task in natural language, execution in a controlled environment with codebase context. Canonical datapoint: ***« Nova's value comes less from the model itself than the systems surrounding it »*** (codebase context, internal practices, safe execution, workflow integration, human review); Nova accounts for **~1 in 12 PRs at Dropbox** today (adoption growing), and extends beyond features to **migrations, flaky-test remediation, bug investigation, dependency updates** (high-toil work). **Measuring product velocity, not code output**: *PR throughput*, a useful signal when coding velocity was the constraint, *« was no longer sufficient »*. A **4-stage** measurement model: ***Fuel*** (are AI tools being used?) → ***Adoption*** (how workflows are changing across teams) → ***Output*** (is AI contributing to production work?) → ***Impact*** (*« improving product velocity and reducing the time it takes to move from idea to customer value »*). Quality signals tracked: **code review turnaround time, first-run test pass rate, defect ratio, rework rate**. *« Quality and trust matter as much as speed »* — the core of the shift: *« moving from local activity metrics toward broader system outcomes »*. **Workflows have to evolve too**: this is *« not just a tooling shift »* but a change of **operating model** — the engineer's role shifts toward *« defining intent, mapping problems, reviewing generated changes, and making higher-context architectural and quality decisions »*. **Enablement** is as crucial as the tool itself (hands-on learning, hackathons, workflow spotlights, bootcamps, peer-led examples); adoption proceeds at varying speeds across teams; *« The goal is not to force every workflow through an agent »* — the goal is to make it *« useful, safe, measurable, and repeatable where it creates meaningful leverage »*. **What we learned**: ***« AI doesn't eliminate bottlenecks in software development, but it does move them »*** (downstream: review, validation, testing, release, prod ops) → optimizing the old bottleneck no longer creates the same leverage. *« The advantage will not come from access to the same foundation models everyone else can use. It will come from the systems built around those models: context, internal tooling, quality controls, and the workflows that connect them together. »* Pressure also builds **upstream** (product & design): structured specs, design clarity, sharper problem framing. Closing: ***« The future of engineering productivity will not be defined solely by who has the best models. It will be defined by who builds the best systems around them »***; *« The real challenge is no longer just generating more code, but building engineering systems that can reliably turn AI-assisted output into valuable experiences for our customers »*. Direct convergence with **Salesforce/Tallapragada** (Effective Output: measuring value, not volume; no speed/quality tradeoff), **Gupta** (token-to-outcome attribution, cost of a completed outcome), **DORA** (beyond throughput), and the shift of the KPI toward **system outcome** (idea→customer value).
**Kazuaki Okumura** — Dropbox (rôle non précisé dans l'article ; le billet reprend une intervention présentée à la conférence **DX Annual 2026** sur la productivité développeur, ce qui suggère un profil engineering leadership / platform, sans confirmation). Publié sur le **Dropbox Tech blog** (dropbox.tech) · rubrique *culture* · le **28 mai 2026**.
Official **Salesforce News** blog post (*Agentic Enterprise* section, *"Pioneering the Agentic Shift Within Salesforce Engineering"* series), published on **May 27, 2026** (6-minute read) by **Srinivas "Srini" Tallapragada**, *President and Chief Engineering and Customer Success Officer* at Salesforce. Direct follow-up to an earlier post (*"How we got our engineers to use AI — without breaking everything"*) which recounted crossing **>90% adoption**. **Pivot thesis**: Salesforce Engineering moved from a world where AI was a useful *copilot* to one where **agentic tools drive the software development lifecycle (SDLC) itself** — writing code, reviewing PRs, generating tests, updating documentation, managing deployments, coordinating work once handled through human handoffs. **Canonical signal decision**: org-wide standardization on **Claude Code** + ***"we removed all token limits"*** — *"remove every last piece of friction between our engineers and the tools that make them faster and more effective"*. **Major empirical result** (April 2026 vs April 2025): work items completed per developer **+50.8%**, PRs merged per developer **+79%**, and above all **Effective Output score** (an ML measure of the **real value of delivered code**, not volume) **+151.3% year over year**. **Flagship use case**: migration of **33 API endpoints** to a cloud-native architecture, estimated at **~231 person-days** (7 per API) the traditional way, completed in **13 days — 18× faster** — via a **rule-based framework built in Claude** (markdown files + reference implementations), with PR feedback continuously fed back into the rule set, **autonomous LLM loops (build, fix, validate)** with no manual intervention, parallelized across isolated environments → **5 PRs**, the largest delivering **21 endpoints with 100% test coverage**. **No speed↔quality tradeoff**: through the **Engineering 360** platform (centralizing engineering data from hundreds of systems), **total incidents drop by 5%** despite the rise in PRs (*"quality doesn't suffer from speed. It benefits from it"*), thanks to **security guardrails and quality standards structurally embedded** in the agentic workflow (Trust as the #1 value). **SDLC overhaul**: once AI is adopted, engineers **tear down and rebuild** workflows (which processes to eliminate? which handoffs are now unnecessary? where does a human still do work an agent could own?). **New engineering craft**: **Claude Code skills** (packaged, reusable capabilities encoding team context, naming conventions, patterns) become a shared, composable **engineering artifact**; **AI Expert Suite** + **Salesforce Foundation Plugins** = an institutionalized, curated skills library (internal benchmark: **higher accuracy and reliability, reduced unnecessary cost**); **subagents & agent teams** parallelize workstreams (*"They describe the outcome, and a set of coordinated agents figures out the steps"*). **What remains hard**: (1) **context management** in long sessions — **CLAUDE.md file quality** varies widely and weighs heavily on output quality; (2) **agentic security** = a fundamentally different model (agents that *act*, not just *suggest* → increased blast radius); (3) **evolving roles** (how do juniors become seniors if AI absorbs entry-level work? role of the designer/PM? the execution unit = scrum team → experiments with 1- or 3-person units). Conclusion: *"It changed what was economically possible"*; the stated ambition is **"the most automated, agentic SDLC in the industry"**. Directly intersects with Gupta (*cost of a completed outcome*, marginal token utility), Greenwald/Sierra (outcome-based pricing), DORA (ROI / cost per feature) and the BFM/Girard debate (token as a value fuel, not a cost to cut).
#Agentic SDLC#agentic SDLC#Claude Code
**Srinivas « Srini » Tallapragada** — *President and Chief Engineering and Customer Success Officer* de **Salesforce**. Plus d'une décennie chez Salesforce · dirige l'ingénierie mondiale de la plateforme unifiée. Auteur de la série *Agentic Enterprise* sur le blog Salesforce News ; ce billet (27 mai 2026) est la **suite** d'un premier opus consacré à l'adoption de l'IA par les milliers d'ingénieurs Salesforce (*« How we got our engineers to use AI — without breaking everything »*). Position d'autorité = **dirigeant exécutif** parlant en son nom et au nom d'une organisation d'ingénierie à grande échelle (donnée terrain à l'échelle d'un hyperscaler SaaS) · avec accès aux métriques internes (Engineering 360, Effective Output).
**Zed** blog post signed by **Franciska Dethlefsen** (head of growth and marketing), published on **May 14, 2026** — the day after Anthropic's announcement — to answer Zed users' questions. **Subject**: starting **June 15**, Anthropic **splits Claude subscription billing into two pools** — one for its **first-party tools** (chat, the official Claude Code CLI), the other for **third-party agent and SDK usage** (anything going through **ACP**, `claude -p`, or a third-party tool). Usage via ACP **then stops drawing from the Pro or Max limits** and switches to a monthly **"Agent SDK" credit**: **$20 for Pro, $100 for Max 5x, $200 for Max 20x**. Once the credit is exhausted, usage continues **at standard API pricing** if overage is enabled — otherwise requests stop until the next cycle. **The figure driving the article**: subscriptions had until then subsidized agentic usage by a factor of **≈15 to 30×** relative to API pricing, and the new credits are billed **at full API rates** — hence *« for anyone using agents heavily, this is a major cost increase »*. **Three options are proposed**, in an order that reveals Zed's position: (1) keep the subscription by running the **official `claude` CLI in a terminal inside Zed** rather than via ACP — *« when the official claude CLI runs in the terminal, it uses your subscription's limits, not the new credit »*; (2) use Zed's built-in agent with the provider of one's choice (Zed-hosted models, API keys, Copilot, local Ollama, DeepSeek); (3) plug in **any ACP agent** — OpenCode, Codex, Factory, Cursor —, several of which still offer rate-limited subscriptions that subsidize heavy usage. **The underlying thesis**, and the real reason for the post: *« ACP is an open protocol… so that your editor is never locked into one provider's pricing decisions »*, with the explicit anticipation that *« this kind of change won't be the last »*. **The post carries an addendum dated June 16, 2026** announcing that **the change is suspended**: ACP, `claude -p`, the Agent SDK, and third-party applications continue to work with subscriptions **as before**, no separate credit to claim, limits unchanged, Anthropic revising its plan with advance notice announced. **The artifact is thus self-contradicting**: its most important content — the reversal — postdates its own publication date by a month.
#Zed#Anthropic#Claude subscription
**Franciska Dethlefsen** — head of growth and marketing chez **Zed Industries**. Le rôle est déterminant pour lire le texte : ce n'est pas un billet d'ingénierie mais une **communication de crise produit** · écrite le lendemain d'une annonce d'un fournisseur dont Zed dépend · à destination d'utilisateurs inquiets. La signature growth/marketing explique la structure (problème → options → réassurance) et le fait que l'argument protocolaire arrive en conclusion plutôt qu'en tête.
Continuous Delivery as the non-negotiable foundation of AI-assisted development — Dave Farley, on his channel *Modern Software Engineering*, argues that without CD, AI is not an accelerator but a trap (theory of constraints and Jevons paradox applied to generated code, ATDD/BDD as a safeguard, deployment pipeline as quality arbiter).
#Continuous Delivery#Generative AI in the SDLC#ATDD (Acceptance Test-Driven Development)
Dave Farley (Modern Software Engineering — YouTube channel)
Manifesto-style article by **Thariq Shihipar** (Engineer & serial entrepreneur, Claude Code team at Anthropic) announcing a **change in the default output format for agents**: replacing **Markdown with HTML**. Thesis: Markdown has been the dominant format between humans and agents (simple, portable, editable, readable) but has become **a bottleneck** as agents produce longer and richer artifacts (specs, plans, reports, code review). Beyond ~100 lines, no one reads a Markdown file anymore. HTML solves six limitations simultaneously: **information density** (tables, CSS, SVG, scripts, canvas, images), **visual clarity** (navigable, mobile-responsive layout), **ease of sharing** (an S3 link directly openable in a browser), **two-way interactivity** (sliders, knobs, "copy as JSON/prompt" buttons to loop back into Claude Code), **native contextual ingestion** (Claude Code reads the codebase + MCP Slack/Linear + git history + Chrome) and **enjoyment** (the author explicitly claims *"it's joyful"*). Five canonical uses detailed: (1) **specs/plans/exploration** in a comparative grid, (2) **PR review** with inline annotated diff, (3) **design & prototypes** with animation sliders, (4) **reports/research/learning** (the author had a prompt-caching explainer generated from git history), (5) **custom throwaway editors** (drag-and-drop of Linear tickets, feature-flag editors, side-by-side prompt-tuner) that produce a re-injectable "copy as markdown/diff/JSON" export. Explicit anti-pattern: *"I'm a little bit afraid that people will read this article and turn it into a /html skill"* — the author **rejects premature skill-ification**, recommending prompting from scratch ("make a HTML file"). Pragmatic FAQ: token cost absorbed by **Opus 4.7**'s 1MM context, 2-4× longer generation, noisy HTML diffs (a real downside), style kept in check via a reference HTML design system.
#HTML#Markdown#output format
Thariq Shihipar (Engineer & serial entrepreneur, équipe Claude Code chez Anthropic — site : thariqs.github.io/html-effectiveness ; X : @trq212)
Editorial by Andrew Ng in The Batch n°352 of May 8, 2026 — **"There Will Be No AI Jobpocalypse"** — which dismantles the narrative of mass unemployment caused by AI, drawing on the **healthy 4.3%** US unemployment rate and robust tech hiring. Ng identifies **three drivers** of the jobpocalypse narrative: **(1) tech incentives** — AI labs benefit from presenting themselves as transformative-disruptive (funding rounds, valuations, talent); **(2) pricing power** — vendors charge **$10,000+/year** to enterprise clients by **anchoring their pricing on the salary of the replaced employee**, rather than on traditional SaaS pricing (per seat / per usage); **(3) corporate messaging** — companies reframe their layoffs as *"AI efficiency"* rather than acknowledging the **pandemic-era overhiring** of 2020-2022. Honest acknowledgment: *"AI disrupts work"*. But Ng flips this into **"AI jobapalooza"** (a play on Lollapalooza) — job creation in AI engineering and adjacent fields with evolving skill sets. Implicit tension with **Amodei** (50% of white-collar jobs eliminated by 2030) — Ng points out, without naming him, that **Anthropic benefits from promoting this narrative** (tech incentives). Published **the same day** as **Wallace-Wells's "AI Populism" NYT Magazine** piece: a perfect mirror reading — Ng = cold economic analysis / Wallace-Wells = popular panic. Pricing-power convergence with **Bain's "$100B cross-system labor"** (same thesis: pricing anchored on salaries).
#Andrew Ng#The Batch#DeepLearning.AI
Andrew Ng (fondateur DeepLearning.AI, Stanford, ex-Google Brain, ex-Baidu, ex-Coursera)
Podcast by Greg Isenberg × Meng To (designer, founder of Design+Code, creator of the products Aura / New Form / Dream Cut) on **`design.md`** — Google's open-source convention, equivalent to `agents.md` / `skills.md` / `soul.md` but **for the design system** (typography, colors, spacing, WebGL/Three.js animations, reveal rules). Central idea: carrying the "**soul of design**" in a markdown file that is handed to an agent (Claude Code, Codex, OpenClaude, Gemini, Stitch, Aura, V0, Lovable, Cursor) to preserve **cross-medium consistency** (web, mobile, Replit slides, Hyperframes/Remotion motion design). Triad taught: **HTML = finished dish, design.md = recipe, skills = ingredients** (typography, lasers, skeuomorphic, 3D skills — 63 in New Form). Major diagnosis: **design drift** on one-shot workflows (`v0`, Lovable, Framer) that start strong then drift into generic output. Meta-message: *taste* is the only remaining **moat** — *"if something looks like another thing, its value drops by 10× to 100×"*. Workflow: **Reference → Design.md → Generate → Inspect → Systemize → Iterate (up to 1000+ prompts) → Remix → Expand → Export**. Critique of **purple gradients** ("you just run") as the generic post-vibe-coding baseline. Meng To claims to have spent ~$500,000 in tokens, run 1,000–10,000 iterations per product, and managed 4 products in parallel solo.
#design.md#Google#design system
Greg Isenberg (host — podcast Late Checkout / The Greg Isenberg Show, 12 mai 2026 livestream workshop ideabrowser.com) ; **Meng To** (guest — designer, fondateur Design+Code 2014, créateur Aura / New Form / Dream Cut, autodidacte parti à 18 ans, dropout, francophone d'origine canadienne)
Interview with Boris Cherny (creator of Claude Code, Anthropic) at a Sequoia event (hosts: Asia, Lauren Reader). Cherny states ***"coding is solved"***: he himself has written **0 lines of code** since late 2025, the model writes **100%**, *"a few dozen PRs/day, 150 PRs in a single day record"*. Account of the genesis of Claude Code (Anthropic Labs incubator late 2024, Mike Krieger in charge of round 2, pre-PMF build *"for the next model"*, a first release that didn't take off, **exponential growth started with Opus 4 in May 2025**, accelerating with each new model 4 → 4.5 → 4.6 → 4.7). Current personal setup: **"most of my work I do from my phone"** (iOS), 5-10 sessions, **"a few hundred agents going, a few thousand at night"**, **`/loop` is the future** (cron + repeat jobs, agents babysitting CI, rebasing PRs, clustering Twitter feedback). **Routines** = the server-side equivalent, running with the laptop closed. SaaS outlook: no apocalypse, but a **reshuffling of Helmer's 7 Powers framework** (switching costs ↓, process power ↓, network effects/scale economies/cornered resources unchanged) and **10× more disruptive startups** over the next 10 years. Pivot analogy: the **Gutenberg press** (10% literacy in the 1400s → 70% over the following centuries, books 100× cheaper within 50 years), *"software will be similarly democratized, but faster than 50 years"* — *"the best person to write accounting software is not an engineer, it's a really good accountant."*
#Boris Cherny#Anthropic#Claude Code
Boris Cherny (créateur de Claude Code, Anthropic) interviewé par Lauren Reader (Sequoia) avec introduction d'Asia (Sequoia).
GitHub repo `techygarg/lattice` that formalizes a framework of **composable skills** for installing an *engineering discipline* into AI coding assistants (Claude Code, Cursor). Distinctive three-tier architecture: **Atoms** (single-principle guardrails: clean code, DDD, security, test quality, design-first), **Molecules** (multi-step workflows composing the atoms: design, implement, refactor, fix, review), **Refiners** (guided interviews producing project-specific standards that customize the atoms' behavior). Operational pipeline `lattice-init` → `design-blueprint` → `code-forge` → `review`, with `refactor-safely` and `bug-fix` as offshoots. Three pivotal principles: *"Skills over prompts"*, *"Composability over monoliths"*, ***"Living context over static config"*** — the `.lattice/` folder grows smarter with every feature cycle. MIT, pure shell, 18 stars / 52 commits, a series of articles on martinfowler.com explaining five *collaboration patterns*. Strong convergence with Vincent *Superpowers* (2026-04-02), Habert *PROJ-AI* (2026-05-05), Wescale *Usine Logicielle Augmentée* (2026-05-03), and — the highest doctrinal convergence with no declared lineage — **Compound Engineering** by Every (Shipper/Klaassen 2025-12-11): isomorphic pipelines (lattice-init→design-blueprint→code-forge→review ↔ ce:brainstorm→ce:plan→ce:work→ce:review), living context layer (`.lattice/` ↔ `docs/plans/+solutions/+brainstorms/`), a shared design-first stance, mandatory review at the end. 2026 *coding agent harness* doctrine converges on a stable vocabulary, without direct influence.
#lattice#techygarg#composable AI skills
techygarg (auteur GitHub, identité réelle non précisée dans le README ; auteur d'une série d'articles publiée sur martinfowler.com).
Google whitepaper (the "Day 1" installment of a series, by Addy Osmani, Shubham Saboo and Sokratis Kartakis) mapping the transformation of the software development lifecycle (SDLC) in the age of coding agents. Thesis: the fundamental shift is not a new language but the move from writing code to **expressing intent**. The document sets out a spectrum ranging from *vibe coding* (prompting and accepting) to *agentic engineering* (AI implements under constraints, tests, and feedback loops designed by humans), with **context engineering** as the central skill, the **software factory** model (the developer's deliverable = the system that produces the code), **harness engineering** (Agent = Model + Harness), and a CapEx/OpEx economic analysis of total cost of ownership.
Interview with Andrej Karpathy (OpenAI co-founder, former Tesla Autopilot) moving from *vibe coding* to *agentic engineering*: December 2025 as the turning point "never felt more behind as a programmer," the Software 1.0/2.0/3.0 taxonomy, the openclaw example (bash script → text to copy-paste into the agent) and MenuGen rendered obsolete by Gemini's Nanobanana, the *verifiability* theory explaining why LLMs are *jagged* (math/code peak, "walk to the car wash 50m away" fails), the distinction between *vibe coding* (raise the floor) and *agentic engineering* (preserve the quality bar), the "animals vs ghosts" metaphor, the overhaul of hiring via agent-versus-agent projects, and the key formula: ***"You can outsource your thinking but you can't outsource your understanding."***
#Andrej Karpathy#vibe coding#agentic engineering
Andrej Karpathy (co-fondateur OpenAI, ex-Tesla Autopilot, créateur du terme "vibe coding")
Manifesto-style X thread by Rohit (@rohit4verse) laying out the *2026 AI engineer roadmap*: a $150k gap between prompt engineer and systems architect, the end of *generic wrappers* "sherlocked by big tech," and 5 portfolio projects ranked by complexity level (offline mobile SLM, self-improving coding agent, multimodal *Cursor for video editors*, privacy-first personal life OS agent, autonomous enterprise workflow agent). Each project describes its *key architectural decisions* (lazy loading, sliding window, sandboxing, scene detection, personal knowledge graph, event-driven multi-agent, audit trail, RBAC, observability). Structuring slogan: *"the replaceable: building wrappers / the unfireable: shipping autonomous systems"*. Injunctive, viral tone typical of X in 2026.
#2026 AI engineer roadmap#Rohit#rohit4verse
Rohit (@rohit4verse) — créateur de contenu IA sur X · vulgarisateur d'architecture et roadmaps de carrière en ingénierie IA.
Post from the **Ahrefs blog** published on **April 28, 2026** by **Ryan Law** (Director of Content Marketing, Ahrefs) describing an in-house **content engineering** system built around **Claude Code**: an editorial pipeline that produces **publish-ready drafts in 6 to 12 minutes**. **Pivot thesis**: ***« AI content is not, by default, good. This process works well because it mirrors our existing human editorial process »*** — quality doesn't come from the model but from the **faithful reproduction of a human editorial process** proven over decades. Architecture: **~23 skill files**, each corresponding to an editorial step (keyword research, topic gap analysis, structural outlining, research compilation, draft generation, formatting), **orchestrated by a master skill `blog-pipeline`** that chains them to produce a complete article. **Seven design principles**: (1) **mimic human workflows** by chaining skills adapted from existing Ahrefs editorial documentation; (2) **output each step separately** for troubleshooting (*« if you get an article at the end of a ten minute run, and it's bad, it's hard to diagnose precisely where and why the process went wrong »* → save intermediate outputs); (3) **create test cases** via Anthropic's `skill-creator` skill to evaluate and improve guidance; (4) **plug in quality data sources** — the **Ahrefs MCP** (keyword metrics, parent topic, long-tail themes, SERP overviews, competitive analysis), competitive analysis and product docs; (5) **front-load human direction** via context parameters enabling editorial guidance; (6) **build interactive previews** in HTML format for review before publication; (7) **allow customization** (each team member can fork and modify the system). **Volume**: ~**15 articles published** and ~**30 articles updated** via this workflow; development started in **February 2026** (the prior process from **August 2025** took several days and manual intervention). **Explicit caveats** (anti-oversell): *« experience matters »* — the process reflects decades of editorial expertise; topic selection focuses on **informational SEO content** the author knows well; Ahrefs **has no plan to "scale" content massively** but maintains an **evergreen library**. Philosophy: automate *« the formulaic parts of work »* to eliminate drudgery and free up time for research, thought leadership, webinars, and system optimization — **not** replace human effort. Canonical reference cited by Pasquale Pillitteri (*Opus 4.8 SEO workflow*) as field proof of the « 6-12 min/draft » gain. Direct convergence with the **skills-over-prompts** doctrine (Lattice, PROJ-AI), **systems around the model** (Dropbox/Okumura), and the use of **HTML as a review artifact** (Shihipar).
**Ryan Law** — Director of Content Marketing chez **Ahrefs**. Praticien senior du content marketing SEO ; le billet est un retour d'expérience personnel (*« How I do… »*) publié sur le **blog Ahrefs** (ahrefs.com/blog) le **28 avril 2026**.
Andrew Ng's editorial in The Batch #350 sets out an **acceleration hierarchy for coding agents** by type of software work: **Frontend (max) > Backend (moderate) > Infrastructure (low) > Research (minimal)**. The rationale rests on implicit *verifiability* (fluency in TypeScript/JavaScript plus an autonomous agent–browser test loop on the frontend) and on the LLMs' blind spots (corner cases / security / DB migrations for backend, opaque network tradeoffs for infra, irreducible hypothesis formation for research). The issue is rounded out by 4 structuring news items: **GLM-5.1 (Z.ai)**, a 754B/40B-active-parameter MIT-licensed model capable of autonomous tasks lasting 8 hours (SWE-Bench Pro leader at 58.4%); **Digit (Agility Robotics) at Schaeffler**, the first industrial deployment of humanoids (5'9"/143lb, $10–25/h vs $20/h for a human); the **anti-data-center revolt** (~$64B blocked May 2024 – March 2025, Maine moratorium on 20MW+ facilities, molotov cocktail at Sam Altman's home); and the **"assistant axis"** (Christina Lu, MATS / Oxford / Anthropic), which reduces persona drift and jailbreaks (Qwen3 32B: 83%→41%; Llama 3.3 70B: 65%→33%) without degrading IFEval/GSM8k/MMLU-Pro/EQ-Bench.
#Andrew Ng#The Batch#DeepLearning.AI
Andrew Ng (édito principal — fondateur DeepLearning.AI, Stanford, ex-Google Brain, ex-Baidu) ; rédaction The Batch (DeepLearning.AI) pour les sections actualités
Synthesis by Addy Osmani (Google, Chrome/Cloud) of the emerging field of *harness engineering*: the equation `agent = model + harness`, the *ratchet* principle ("every mistake becomes a rule"), the HumanLayer "skill issue" reframe, Terminal Bench evidence (Top 30 → Top 5 from a harness change alone), the layered Claude Code architecture, Anthropic's "harnesses don't shrink, they move" vision, and Harness-as-a-Service (Claude Agent SDK, Codex SDK, OpenAI Agents SDK). Pivot article that consolidates Trivedy, HumanLayer, Anthropic, and Böckeler into a doctrine.
#harness engineering#agent harness#Addy Osmani
Addy Osmani (Software Engineer at Google, Cloud + Gemini)
Public update from Darragh Curran (R&D, Intercom) nine months after his commitment to double R&D productivity in 12 months through AI. Result: **3x achieved in 16 months, with no signs of plateauing**. Quantified data from a 500-person R&D organization / 8.5M lines of code: **93.6% of PRs are agent-driven**, **19.2% AI-approved** (target >50%), cost/PR **-50%**, defect backlog **-54%**, time-to-shipping **-39%**, breaking-changes downtime **-35%**, top 5% of performers at **6x the median PR throughput**, **497 autonomous PRs** in the first 4 weeks, **153 contributors / 267 specialized skills** in a private *Skills-Based Plugin Architecture*. Curran declares ***"All technical work is becoming agent-first. This is the top priority for R&D."*** Pivotal article of the *agent-first organization* dossier, comparable only to Stripe Minions and StrongDM in the 2026 corpus.
#Darragh Curran#Intercom#Fin Ideas
Darragh Curran (R&D leader, Intercom — publication via Fin Ideas, plateforme média Intercom).
Anthropic Interviewer - AI Workforce Study - Professional Perspectives - AI Adoption - Workplace Transformation - Creative Industries - Scientific Research - Economic Impact - Sociological Research - AI Sentiment Analysis
Anthropic Research - AI Work Transformation - Claude Code Impact - Software Engineering - AI Adoption - Productivity Study - Workplace Evolution - AI Collaboration - Skills Development - Future of Work
#Anthropic#AI Transformation#Workplace Impact
Anthropic Research Team (132 engineers and researchers surveyed, 53 in-depth interviews conducted)
Cat Wu and Boris Cherny (Anthropic) explain how to use Claude Code like its creators: antfooding, plan mode, subagents, hooks, and extensibility — Every's AI & I podcast
#Claude Code#Cat Wu#Boris Cherny
Rhea Purohit (interviewer: Dan Shipper) · Cat Wu · Boris Cherny