Archive · 38 issues
The archive
Every issue, nothing gated. The archive is the sample: if the last two weeks read well to you, the email will too.
- Daily
Most of a harness-evolution gain is overfitting and extra compute
Plus: Monzo's 1,800 agent tasks a day, and a credential the agent thought was yours. 5 min.
Oct 22, 2026 · 5 min

- Daily
Claude Code's Bash tool changed 12% of calls carrying code or escapes
Plus: UndoBench, a Copilot CLI secret leak, and what Codemode is for. 5 min.
Oct 07, 2026 · 5 min

- Daily
Branch steering breaks the Dual-LLM guarantee for computer-use agents
Plus: swarms borrowing public URL scanners, pipelines that go wrong under replay. 5 min.
Oct 06, 2026 · 5 min

- Daily
VS Code rebuilt its release process around agents and now ships weekly
Plus: an agent that writes its own tools, and a support bot at 80%. 5 min.
Oct 05, 2026 · 5 min

- Daily
Six ways an agent harness can run something other than what you approved
Plus: elaborate harnesses lose to a plain coding agent. 5 min.
Oct 02, 2026 · 5 min

- Daily
A forged chat-template marker loses most of its authority as ordinary subwords
Plus: 5 papers, 3 builder posts, 9 quick links. 5 min.
Oct 01, 2026 · 5 min

- Daily
Agent context compression can cut tokens by two thirds and still run slower
Plus: a rescored injection benchmark, Letta's self-written workflows. 5 min.
Sep 30, 2026 · 5 min

- Daily
AISI finds GPT-6 Astra attacking out of scope during a cyber evaluation
Plus: five papers, METR's per-action monitor, and nine quick links. 5 min.
Sep 29, 2026 · 5 min

- Daily
Agent-written pull requests match human ones on revert rates, and fail differently
Plus: two papers pricing what agent context costs to serve and to keep. 5 min.
Sep 28, 2026 · 5 min

- Daily
Forged control tokens blank an agent's reasoning while the tool call still fires
Plus: a completion gate you can prove, and 8 tools where 83 were. 5 min.
Sep 25, 2026 · 5 min

- Daily
Two MemOS packages shipped credential stealers into the agent memory layer
Plus: an agent that spent eight days rewriting itself, and an MCP hijack that rides tool metadata. 5 min.
Sep 24, 2026 · 5 min

- Daily
Dormant prompt injections land on nine production agents where direct orders fail
Plus: what context compression actually saves, and 383 self-written rules on trial. 5 min.
Sep 23, 2026 · 5 min

- Daily
Approve one operation, run another: the binding failure in shipped agent products
Plus: Benchling's DNS-tight sandbox, Shopify's Swift rewrite, nine skill techniques. 5 min.
Sep 22, 2026 · 5 min

- Daily
Giving an agent the decoder's confidence never beat a plain deterministic gate
Plus: three agent architectures priced against the same task, and nine quick links. 5 min.
Sep 21, 2026 · 5 min

- Daily
Split an MCP injection across two channels and resistant models leak at 100%
Plus: reward hacking at 57.2% of rollouts, and Exa searches the web by date. 5 min.
Sep 18, 2026 · 5 min

- Daily
Every audited chat tokenizer lets prompt text forge control tokens
Plus: 91% of audited vibe-coded deployments shipped a hole. 5 min.
Sep 17, 2026 · 5 min

- Daily
A shell beats a typed tool catalog, and agents barely report their work
Plus: an MCP registry census, four AI Engineer talks, one very late transcription. 5 min.
Sep 16, 2026 · 5 min

- Daily
An agent deleted an AML control, and benchmark scaffolds do the model's work
Plus: OpenAI's 10,000-agent proof, a 297-iteration model loop, MCP merges Skills. 6 min.
Sep 14, 2026 · 6 min

- Daily
Per-phase model routing cuts agent cost, RASER on Slurm, Amp steers mid-run
Plus: 2 papers, a context-mode split, and OpenAI's Navier-Stokes claim. 5 min.
Sep 09, 2026 · 5 min

- Daily
Prefix caching changes agent runs, and cross-family reviewers beat self-review
Plus: a harness that moves solve rates 4x, and 883 commits to nowhere. 6 min.
Sep 08, 2026 · 6 min

Browse by topic



















