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/`.
Internal research report dated **August 12, 2026** (in *What? — So What? — Now What?* format, investigation conducted August 11-12) on a simple question: are the **desktop** applications of ChatGPT and Claude better than their **web** versions? The answer comes in two parts. **(A) A solid, well-sourced qualitative consensus exists.** The starting point is indisputable: desktop and web call exactly the same cloud models, the application being merely an interface to the service — the gain therefore lies entirely in the application shell (access latency, stability during long sessions, memory footprint, system integrations, workflow fluidity). What genuinely distinguishes desktop, confirmed: on the OpenAI side, a global shortcut (Option/Alt + Space), a *companion window* that always stays on top, native screenshots, and since July 2026 the **Codex/Work** agentic capability built into the app; on the Anthropic side, **Quick Entry** (macOS), **Desktop Extensions** (installing a local **MCP** server becomes *"as simple as clicking a button"*), access to local files, **Cowork** and **Computer Use** (Accessibility permissions and screen recording). The web retains two confirmed strengths: multiple tabs/threads, and universality without a client to install. **(B) Nearly all the figures circulating to support this consensus do not withstand verification.** The report's critical audit (§1.5) classifies **unconfirmed** seven widely repeated numerical claims: the *cold start* "2-3 s vs 8-12 s" (the only trace being an anecdotal *"loads in about 3 seconds"* on Substack); RAM usage "200-700 MB vs 1.2-2 GB," attributed to an "Alibaba Product Insights" whose pages return **404**; an untraceable glitch rate and session retention figure; a "Claude +10-20% end-to-end" attributed to **Skywork**, which had in fact benchmarked its own Windows agent rather than Claude against the web; an untraceable "Cosmo Edge" source; unconfirmed Zenken AI citations; and two unauthenticated X posts with no URL. The counter-signal is documented with the same rigor: Yuri Dvoinos describes a Claude Desktop app that *"makes me want to throw my laptop out the window"* — 68% CPU usage, input lag on a MacBook Pro — and the report notes that both apps are **Electron** builds with native layers. Hence its formulation: *the desktop advantage is a promise of implementation, not a law of nature.* **The "So What"**: since the model has become the common denominator, the interface becomes the battleground — the **Codex + ChatGPT** merger of July 9, 2026 and the Cowork/Computer Use tandem tell the same story, *"the desktop app is no longer a chat client, it's an agent runtime with access to the machine."* Three consequences: the gain is a **friction** gain, not a power gain; for a CIO, desktop **shifts the trust boundary** — Computer Use requires sensitive system permissions and the Codex merger places code execution, browser, and connectors within *"one expanded trust boundary,"* whereas the browser remains governable via SSO, DLP, and CASB; and for anyone publishing, the fragility of the figures is itself the story. **The "Now What"** delivers individual switching criteria, a CIO checklist (inventory permissions, disable Computer Use and Cowork by default, scope which MCP extensions are authorized, organize distribution and updates — on Linux, outside the apt repository, Claude Desktop does not update itself) and an editorial directive: cite only confirmed verbatims and dates.
#ChatGPT Desktop#Claude Desktop#web version
**Deep Research Veille Interne** — rapport non signé · produit par une enquête sourcée menée les **11-12 août 2026** et rendu le 12.
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**.
News article analyzed, published on **VentureBeat** on **August 11, 2026** by **Michael Nuñez**, based on an **exclusive interview with Timothée Lacroix**, co-founder and CTO of **Mistral AI**, conducted ahead of the announcement, ~2,000 words. Mistral is expanding its infrastructure offering in three parts: **Mistral Regional Endpoints** in general availability (pinning inference and its associated processing to Europe or the United States), a **Priority Tier** in public preview (committed service levels, custom quotas, availability SLA), and a **coalition of European enterprises** whose multi-year commitments are meant to fund **200 MW by the end of 2027** and **1 GW by the end of 2030**. The vehicle is called the **European Compute Unit (ECU)**: a claim on capacity built by Mistral, fungible across inference, training, model adaptation, or managed Kubernetes, over a targeted five-year horizon. Lacroix describes the mechanism bluntly — *"The whole point of compute units is to have commitment"* — and, on early exit: *"There is no getting out."* The article scales the ambition: Mistral states it operates *"less than 200 MW"* and details three sites totaling **77 MW** (44 MW near Paris, 23 MW in Sweden with EcoDataCenter, 10 MW in Les Ulis); **Epoch AI** puts the initial capex for a one-gigawatt AI datacenter at **~$38B**, and **Goldman Sachs Research** puts next-generation facilities at **$15-20M/MW excluding chips**, against the **~$4B** Mistral has raised in total (PitchBook). Added to this is a decision that *"is likely to raise a few eyebrows among sovereignty purists"*: Mistral is starting to **host third-party open models**, beginning with **GLM-5.2** from **Z.ai**, a Chinese lab — *"It's a great model. Everyone loves it. It's open-weight, so there was no good reason for us not to do it."* The article digs into the fine print of Mistral's documentation, which mentions *"limited, controlled transfers"* to subcontractors outside the region; pressed for detail, Lacroix points to **tool calls**, web search in particular, and states that **gating is the feature, not the bug**. The author's framing: *"full regional control is available, but the moment an AI agent reaches out to the open web, sovereignty becomes a configuration decision, not a default."* Two dependencies remain: **GPUs** come from Nvidia, and **Microsoft** — anchor tenant of Mistral's European datacenters since July — is presented as what de-risks the buildout.
#Mistral AI#digital sovereignty#AI sovereignty
**Michael Nuñez** — journaliste **VentureBeat** · couvre l'IA et l'infrastructure ; déjà présent au corpus. L'article est bâti sur un **entretien exclusif avec Timothée Lacroix** · cofondateur et CTO de Mistral AI · conduit **avant l'annonce** · et fait suite à un entretien de juin avec le même interlocuteur. Publié le **11 août 2026**.
Long-form article published on **X** on **August 11, 2026** by **Jesse Zhang**, CEO of **Decagon** (customer-service AI agents), under a dilemma-shaped title — *« To FDE, or not to FDE? »* — devoted to the **Forward Deployed Engineer**, which has become *« the answer to almost every hard question in AI go-to-market »*. Starting observation: Anthropic and OpenAI have built enterprise deployment arms explicitly modeled on Palantir, *« every seed-stage company »* advertises an FDE offering, and job postings for the title are said to be up several hundred percent in a year. **(A) The Palantir genealogy** supplies the framework: **Shyam Sankar**'s (CTO) formula, *« FDEs eat pain and excrete product »*, and **Joe Lonsdale**'s reminder that Palantir spent nearly two decades being called a *« glorified consultancy »* on the basis of an accurate observation. **Gotham**'s bespoke deployments (CIA, NSA, military intelligence) were encoded into platform primitives — ontology, object models, permissions, workflow engines, provenance tracing — which became **Foundry**, then Apollo and AIP; standardization pushed gross margin into the 80% range and Palantir moved from an FDE motion to account-based selling, with many FDEs migrating into core engineering. *« The pain was the input to the product, not a cost of sale. »* **(B) The criterion proposed** is not to give up on FDEs but to know when to stop: go early, then ask whether one is still **discovering** — *« The trap is not starting. It's not stopping. »* **(C) A distinction few make: FDE ≠ implementation.** *« Building that integration into their ticketing system »* is real work, but it is execution against a known spec, not discovery of an unknown one; conflating the two *« is how a company convinces itself that a growing services org is a product investment »*. Closing line: *« If your FDEs are eating pain and excreting more pain, you don't have an FDE team. You have a services business. »* Two figures are put forward about Decagon — *« two-thirds of deployment work is now done autonomously via Duet »* and *« a few days on average to launch the first AOP, even for large banks, airlines, telcos »* — without the "deployment work" denominator being defined or the AOP acronym spelled out.
#Forward Deployed Engineer#FDE#engineer embedded with the client
**Jesse Zhang** — cofondateur et **CEO de Decagon** (agents IA de service client, San Francisco) · 85 000 abonnés sur X · site personnel `jessezhang.org`. Il cite son cofondateur **Ashwin Sreenivas** · **ex-Palantir** · d'où la profondeur du récit Palantir. Publié le **11 août 2026**.
Doctrinal manifesto published on **meta.com** on **August 10, 2026**, signed with only a first name (*"– Mark"*) by **Mark Zuckerberg**, under the title *"The Future is for Everyone: The Path to a Positive AI Future"*, ~6,500 words. Three principles are announced from the outset: individual empowerment as a source of prosperity, invention as the primary purpose of superintelligence, balance of power as the foundation of safety. **(A) The central argument is a political argument**, stated as a short chain: *"Humanity is not a monoculture"* — people's values encode opposing trade-offs, no technical solution can align simultaneously with conflicting interests, so any singular superintelligence would have to prioritize certain values over others and would thereby be incapable of being benevolent toward everyone. Hence the formula: *"There is no such thing as a singular benevolent superintelligence."* Safety is reframed as a problem of power distribution, illustrated by a thought experiment repeated three times (a single superintelligent lawyer versus everyone having one; the same for cybersecurity, then for business). **(B) A redefinition of alignment**: *"Solving alignment is necessary for billions of people to adopt personal superintelligence agents. But it also implies that if we reach a state where billions of people are using and scrutinizing personal superintelligence agents, then we will have solved alignment with their interests."* The corollary targets the rest of the industry without naming it: *"the most dangerous scenario would be leading labs training powerful models and keeping them for themselves."* **(C) Datable commitments**: a **fully private** mode where *"even Meta"* cannot see or grant access (a WhatsApp analogy); **free** versions for billions of people paired with a **dynamic bidding mechanism** for paid compute; the announced **resumption** of open source releases — *"we will soon resume releasing some open source models"*; and a structure giving the **independent board** the power to approve release safety criteria and verify each release's compliance, with the author acknowledging that Meta is a founder-controlled company. **(D) Two public-policy proposals**, repeated three times: that labs share **intermediate training checkpoints** and engineers with the government rather than an end-of-cycle review, and that the **physical production** of dangerous materials be regulated rather than the spread of knowledge. The text's sourcing is nearly nonexistent.
#Mark Zuckerberg#Meta#Meta Superintelligence Labs
**Mark Zuckerberg** — fondateur et PDG de **Meta**. Texte signé du seul prénom (*« – Mark »*) · publié le **10 août 2026** sur un domaine dédié de meta.com. La signature n'est pas « Meta » · et l'alternance des pronoms est régulière : **« we » pour les engagements de l'entreprise** (*« we will offer free versions »*, *« Meta is implementing a governance structure »*) · **« I » pour les affirmations normatives ou contestables** (*« I think this view of alignment is fundamentally flawed »*, *« I propose that companies developing frontier AI should… »*, *« My honest guess, and it is a guess »*). Les engagements produits et de gouvernance sont au « nous » · les propositions de politique publique au « je ».
A watch note by **Didier Girard** published on **X** on **August 7, 2026**, which reads the launch of **Shieldstral 1.0 3B** (Mistral AI, August 4, 2026) not as a product release but as **the production deployment of a doctrine**. Starting point: on **May 13, 2026**, before the National Assembly's commission of inquiry into digital vulnerabilities, **Arthur Mensch** refused any oversight role for Mistral over the end use of its models — *"we do not have democratic legitimacy"* — explicitly rejecting **Anthropic**'s stance. Less than three months later, Mistral releases a **moderation model**. The author dismisses the apparent contradiction: **Shieldstral carries no taxonomy of the licit and the illicit**, it answers a **question the user writes**. **The mechanism is the heart of the note**: a three-part prompt (context + severity / a single closed question / the content to be judged), a `yes` or `no` response, and the **softmax over these two tokens** produces a continuous score between 0 and 1. **The moderation policy is not in the weights, it is read at inference time** — whereas **Llama Guard 4** embeds the MLCommons taxonomy fixed at training time, Shieldstral reads yours in natural language, modifiable **without retraining**. The technical report (**arXiv:2607.25857**, July 28, 2026) quantifies the cost of this choice: fine-tuning on public data alone = **61.1% F1** on policy adaptability; **4.4 million contrastive pairs** generated by an LLM (the same content rewritten to violate a policy but not its sibling policy) = **+23.3 points**; **91.3%** after merging three checkpoints. Characteristics: **3.8B actual parameters** (the "3B" in the name rounds down), **Ministral 3** base + **Pixtral** vision encoder, **12 languages**, **16 GB of VRAM in BF16**, **Apache 2.0**. Text performance: **84.9% average F1**, on par with **GPT-OSS-Safeguard-20B** (seven times larger), ahead of **Qwen3Guard-8B** (84.0) and far ahead of **LlamaGuard-4-12B** (69.1). **A caveat raised by the author himself**: *all these figures come from Mistral, on test sets selected by Mistral, and no third-party evaluation existed as of August 6*. The note's structuring thesis is an **opposition of topologies**: at **Anthropic**, the guardrail lives **in the weights** and the publisher arbitrates who is exempt from it (**Claude Fable 5** public with safety measures / **Claude Mythos 5** without, reserved for approved cyberdefenders of **Project Glasswing**, June 9, 2026); at **Mistral**, the guardrail **sits outside the model** — a separate, open, self-hostable component, whose policy belongs to the deployer. Explicit customer alignment (ministry of the Armed Forces, BNP Paribas, French and Luxembourg government administrations). The note closes on a **setback documented in three points**: **auditability** (binary output, no reasoning trace, while the deployer inherits the burden of justification under an AI Act audit), **robustness** (the first chapter of Voltaire's *Treatise on Tolerance* classified as "calls for violence" by a tester on the Hacker News thread — a mention/endorsement confusion), **availability** (as of August 6: no billed endpoint on La Plateforme, no official Ollama). Three deployment rules to close.
#Shieldstral#Shieldstral 1.0 3B#Mistral AI
**Didier Girard** — auteur de la note · publiée sur son compte X. Écrit ici en **analyste de doctrine industrielle** plutôt qu'en testeur : il n'a pas déployé le modèle · il croise une **audition parlementaire** (Mensch, 13 mai) · un **lancement produit** (Shieldstral, 4 août) · un **rapport technique** (arXiv, 28 juillet) et un **contre-exemple concurrent** (Anthropic, 9 juin) pour montrer qu'ils forment une position cohérente. Deux marqueurs de posture : il **borne explicitement la valeur des chiffres** qu'il cite (aucune évaluation tierce) et il **termine par des règles opérationnelles** — l'analyse doit sortir avec sa traduction en décisions de déploiement.
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`.
Trade-press brief (**Payments Dive**, *Dive Brief* format, **August 6, 2026**) covering **Block**'s quarterly earnings release: the company has already rolled out several AI tools to its customers — **Moneybot** (Cash App) and **Managerbot** (Square) — and has not yet decided how to charge for them. **Jack Dorsey** on the analyst call: *"We're in a fortunate position where we can experiment with a number of models, and then choose the right one that's going to align all of our incentives with our customers."* **The financial backdrop illuminates that stance.** Six months earlier, Block had laid off roughly **4,000 people, about 40% of its workforce**, in a reorganization explicitly framed around AI. In Q2 2026: gross profit **up 25% to $3.2B**, revenue **up 10% to $6.62B**, but **net income at $89M, down 83%** year over year due to severance costs closing out the restructuring; 2026 guidance was raised. AI's value, then, is being captured through the cost structure before it is captured through price. **The heaviest fact sits in the middle of the brief**, drawn from the shareholder letter: *"Starting in June, agentic AI helped write and review nearly all of our production code changes"* — writing **and** reviewing nearly all production code changes, at a publicly traded payments company, six months after cutting 40% of the workforce. A self-reported claim to investors, with no definition of *"nearly all"* or of what *"review"* covers. **The tooling**: **Goose**, an internal system built two years earlier, described as model-agnostic (it plugs in different commercial models for employees); **Buzz**, launched the previous month for *"agent collaboration, communication, and code repositories."* **On the customer side**: Moneybot monitors Cash App user activity and surfaces accounts, balances, and transactions — over **one million weekly active accounts**; Managerbot runs automated marketing, margin analysis, and suggests *"operational fixes"* to Square merchants. **Evercore ISI** analysts list four monetization paths — SaaS bundles, direct subscriptions, enterprise offerings, usage-based pricing — **none of them tied to outcomes**. Stated order of priority: **product quality → distribution → adoption → pricing model**. Two distribution facts round out the picture: Square is rolling into **Google Maps** with a *"conversational AI experience,"* described as *"the first step in a broader partnership between Square and Google"*; and the **Tags** payment device (keychain and NFC chip wands) shows **three million people on the waitlist**. Analyst quotes: William Blair (*"Block epitomizes the secular shift toward tech-forward digital finance firms"*) and Bank of America on the *"post-reset operating model."*
#Block#Jack Dorsey#Cash App
**Justin Bachman** — Senior Reporter · **Payments Dive** (groupe Industry Dive). Journaliste sectoriel paiements ; signe ici un **Dive Brief** · format court en deux temps (*Dive Brief* = les faits du jour, *Dive Insight* = le contexte) qui compile une conférence de résultats · une lettre aux actionnaires · un communiqué et trois notes d'analystes.
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.
Issue of the **Engineering Enablement** newsletter from **DX** signed by **Brian Houck**, published on **August 5, 2026**, approximately **13,000 characters**. The text opens with a quantified observation: at **Meta**, significant lines per diff merged by a human increased by **106%** in one year and diffs per developer per month by **51%**, with more than **80%** of this growth coming from agentic AI, while the share of diffs reviewed within 24 hours declines; DX's analysis for its part measures a median pull request size up **64%**, while developers want to devote about **7%** of their time to review. (A) Before handing review over to AI, Houck reframes the usage question: *"What problem was code review solving before AI arrived?"* If the answer is defect detection, full automation becomes inevitable; the **Bacchelli & Bird** study (*Modern Code Review*, ICSE 2013) shows that defect-related comments account for only **14%** of comments actually written. (B) Part of the burden is *self-inflicted*: the 2016 Microsoft study of **911 developers** already identified turnaround time, review size, and understanding the motivation for the change as the top three obstacles, and only **26%** still wrote a detailed description. (1) **RADAR** at Meta serves as the reference deployment for risk stratification. (2) Three ordered moves: fix the fundamentals, design AI around human judgment, protect what review invisibly produces. The corpus already includes [[monperrus-end-of-code-review-agents-supersede-2026-06-11]], which pushes automation to the point of replacement, and [[laycock-thoughtworks-reviewing-all-this-code-2026-09-02]], the direct response to this text.
#code review#code review#review queue
Brian Houck · signataire de l'édition · sur la newsletter Engineering Enablement de DX ; fonction non affichée par la page.
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.
Product announcement published on the **Cloudflare** blog on **August 4, 2026** by **Will Papper**, as part of **Agents Week**: **Cloudflare Wallets**, presented as *"the programmable wallet for the agentic Internet"*. **The problem stated** is precise and well chosen: an agent that wants to try an API has to go through a login page **designed for humans**, have a human add a payment method, generate an API key, then figure out how to call the service. Two structural gaps explain this — *"Agents do not have a stable identifier to sign up for an API, and they do not have a native way to pay for APIs"* — with the consequence that *"AI agents often give up on these tasks entirely, kicking registration, payment methods, and API key generation back to humans"*. **The proposed architecture comes down to two wallet types**: **Account Wallets**, intended for humans who own a Cloudflare account (fund, delegate, withdraw), and **Virtual Wallets**, intended for agents, **operating via API key** and whose spending cap is **set by the account holder**. The announced guardrails are explicit: **allocation, allow list, maximum amount per transaction**. **The payment rail is the x402 protocol** (payments attached to HTTP requests) and the currency is **stablecoin** — which places the offering in a distinct camp from schemes built on card networks. **The most interesting argument is counterintuitive and central**: *"These limits may seem like constraints, but counterintuitively they give agents more freedom. If an agent is responsible for $10, you can worry less about its spending than if it is responsible for $1,000."* → **the cap is not what constrains autonomy, it is what makes it acceptable.** **Second component, more strategic than the first**: identity, via a **`cloudflare.pay`** namespace — a research agent could live at `research.example.cloudflare.pay`, giving the merchant certainty that it is talking to the agent of an identified organization. Cloudflare claims a deliberately minimal ambition (*"a human-readable identifier for a not-very-readable keypair, similar to the URL and IP-address pairings used in DNS"*), built on its existing building blocks (**Turnstile**, Bot Management, **Web Bot Auth** and its keypairs), and states its intent to adopt the schemes of the **x402 Foundation** as they emerge. **A decisive caveat about the status of the text**: **almost everything is in the future tense**. What exists on the day of the announcement is the **reservation of a handle**; payments, Virtual Wallets, guardrails, and the ramps for accessing funds are announced (*"Soon, you will be able to…"*). This is a **staking of position on a namespace**, more than a service going live.
#Cloudflare Wallets#agentic commerce#Agents Week
**Will Papper** — auteur de l'annonce sur le blog Cloudflare (lecture annoncée : 8 minutes). Publication rattachée à l'**Agents Week** de Cloudflare et étiquetée *Agents Week · AI · AI Bots · Developer Platform · Developers · Payments · Product News · x402*.
**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.
In-depth op-ed published on **sfeir.com** on August 1, 2026, authored by **SFEIR** (the firm's editorial voice). It brings together **two July 2026 publications** with opposite methodologies — the preregistered field experiment **"The Cybernetic Teammate"** at **Procter & Gamble** (Dell'Acqua, Ayoubi, Lifshitz, Sadun, **Ethan Mollick** et al., *Organization Science* 37(4), 2026) and the first report in **OpenAI Economic Research**'s **"Work at the Frontier"** series (Jul. 27, 2026, >800,000 messages from US ChatGPT users) — into a single thesis: *"generative AI doesn't just speed up existing work, it redistributes who does what."* The architecture unfolds in four stages: **the mechanism** (P&G: AI acts as a *boundary-spanning* device, erasing functional silos — an individual + AI reaches the level of a pair without AI, **+0.37 σ**), **the scale** (OpenAI: **43.5%** of profession-specific messages fall outside the user's own profession), **the agenda** (Mollick: the walls are thinning, the division of labor must be rethought, and well-orchestrated recomposition "pays off handsomely"), then **the firm's response** — **Skill Based Organisation (SBO)**, adopted at SFEIR under the impetus of **Rosalie Zandona** (VP People & Culture): **actually operational skill** replaces the job description as the unit of organization (**up to 13 skills identified per role**), shifting from a **status-based identity** ("I am a manager") to an **operational identity** ("I know how to design complex architectures"). The rhetorical move is proof by internal example: *"we made the shift in-house before recommending it."* **Three caveats are noted**: the SBO shift dates back to **February 2026**, hence *predating* the diagnosis it is supposed to resolve (the argumentative order reverses the chronological order); **nothing in the data demonstrates** that a skill-based organization absorbs crossover better than a role-based one (an untested design hypothesis); the P&G result has been circulating **since March 2025** (NBER w33641) — the "a few weeks earlier" applies to the peer-reviewed publication, not to the result itself.
#Skill Based Organisation#SBO#skill-based organization
**SFEIR** — ESN française « AI Only » (~850 ingénieurs, 8 agences France & Benelux). Voix éditoriale du cabinet (byline « SFEIR »).