102 wires and counting

$ follow Anthropic

Keep up with Anthropic in about 3 minutes a day: what actually shipped — the commits, pull requests, releases, and security advisories that matter.

or

fair warning: these emails are deeply technical. diffs, version numbers, CVEs, benchmark deltas. if that's not your idea of a good morning, this isn't your newsletter.

One email a day. Unsubscribe in one click.

$ status

wire 2026-08-29
stories 10

© 2026 RepoJournal Home Showcase Explore How it works Privacy

$ the-wire · showcase

Claude cookbook drops unverified cost claims, formal-math tightens authorship and CI

By RepoJournal · Filed · About Anthropic

Anthropic pulled cost and latency ratios from a managed-agents cookbook after they went unverified, while formal-math standardized authorship and CI for its Palomar submissions.

The claude-cookbooks repo removed the "roughly 2.5x cheaper and 3x faster" claims from CMA_plan_big_execute_small [1], leaving the measured token split (84-98% at the worker rate) intact. This is a prose-only change, with no code or output cells altered [2]. In formal-math, the zeta23 project now presents a consistent authorship account, stating that all Lean code in the repository, including Zeta23/LinAlg/, was written by Claude, resolving a discrepancy flagged by Palomar's automated review [3]. The project also lays out as a Palomar template [4], and per-project CI now builds and runs the Palomar comparator check only on changed projects [5]. Separately, the claude-plugins-official repo promoted scandit-sdk from community to official, adding 86 skills for barcode/ID/label capture across 11 platforms [6].

Action items

References

  1. [1] docs(managed_agents): remove cost and latency ratios from CMA_plan_big_execute_small (#851) ↗ anthropics/claude-cookbooks
  2. [2] docs(managed_agents): remove cost and latency ratios from CMA_plan_big_execute_small ↗ anthropics/claude-cookbooks
  3. [3] zeta23: consistent formalization-authorship account; de-duplicate CI ↗ anthropics/formal-math
  4. [4] Lay out zeta23 as a Palomar template project ↗ anthropics/formal-math
  5. [5] Add per-project CI: build + Palomar comparator check (changed projects only) ↗ anthropics/formal-math
  6. [6] Add scandit-sdk plugin (community → official promotion) (#5634) ↗ anthropics/claude-plugins-official

Quick answers

What shipped in Anthropic on August 29, 2026?
Anthropic pulled cost and latency ratios from a managed-agents cookbook after they went unverified, while formal-math standardized authorship and CI for its Palomar submissions. In total, 5 commits and 5 pull requests landed.
Who contributed to Anthropic on August 29, 2026?
4 developers shipped this update, including mattmccarley-ant, rfurman, jleder3, and abibbs-ant.
What were the notable Anthropic updates?
docs(managed_agents): remove cost and latency ratios from CMA_plan_big_execute_small (#851), docs(managed_agents): remove cost and latency ratios from CMA_plan_big_execute_small, and zeta23: consistent formalization-authorship account; de-duplicate CI.