Latest signal November 19, 2026 · afternoon edition
On November 18 the control points of an agent moved out of the prompt and into the platform, as Microsoft put versioned tools behind one MCP endpoint and outbound traffic behind fail-closed rules while LangSmith tied coding-agent spend to the pull request it produced.
01 / The wire
Recent briefings
-
November 19, 2026 · afternoon
On November 18 the control points of an agent moved out of the prompt and into the platform, as Microsoft put versioned tools behind one MCP endpoint and outbound traffic behind fail-closed rules while LangSmith tied coding-agent spend to the pull request it produced.
-
November 19, 2026 · morning
On November 18 agents moved from asking to staying connected, as GitHub put event streams into gh and added eight MCP connectors and Claude Code let MCP-backed policy hooks block on failure, so what happens when a check breaks is now the setting to audit.
-
November 18, 2026 · afternoon
On November 18 several research teams, Wiz among them crediting an AI harness, found critical NGINX bugs independently while OpenAI showed how it fences its own coding agent, and both stories point at the same gap, code that runs outside the fence.
-
November 18, 2026 · morning
On November 17 finding and doing got cheaper for agents, as a Codex-assisted audit surfaced a critical Java memory bug in an hour and Copilot gained a real browser, while the scarce work moved to checking what agents actually did.
-
November 17, 2026 · afternoon
On November 17 the unit of comparison for coding agents shifted from the model to the model plus its harness, as Runta measured harnesses changing cost per task severalfold with the model held fixed, Microsoft-owned GitHub added harness comparisons and Copilot Evals, and OpenAI and Microsoft pitched their new models on cost per task alongside capability.
-
November 17, 2026 · morning
On November 16 the rules around coding agents moved toward sources the work cannot rewrite, as the Claude Code team described review rules loaded from the base branch and GitHub let admins block standalone Copilot customizations, while the Optimizely compromise showed what a trusted but writable script costs.
-
November 16, 2026 · afternoon
November 16 launches moved the check outside the agent, as OpenAI signed outside assessors for selected training runs, Harmonic took proofs to Rust, GitHub scaled AI Scan to twenty repositories at a time, and CVS published a kit inside OpenAI's healthcare announcement that turns agent runs into code a person can read.
-
November 16, 2026 · morning
Alibaba and Tencent sold their November 16 models on the cost of a finished task instead of the price of a token, and today's trending agent tools measure work the same way, but every per-task number on the table is the publisher's own.
02 / Under the surface
Latest analysis
-
rust-anything Checks Whether Your AI Agent's Rust Rewrite Actually Works
Rust-anything makes the original program's recorded behavior the spec for an agent-written Rust port, so a pass means it matches every…
-
Microsoft Foundry's Crash Recovery Reruns Your AI Agent. Make Sure It Doesn't Email Twice.
Foundry hosted agents now recover from crashes by rerunning the handler from the top, which makes every side effect at-least-once, so the…
-
Claude Code Safety Hooks Fail Open by Default. One Setting Makes Them Actually Block.
A Claude Code policy hook that crashes, times out or cannot start lets the action through by default, so every hook you rely on as a gate…
-
Bitwarden Agent Access: You Don't Have to Give AI Agents Your Whole Vault Anymore
Bitwarden Agent Access narrows what an agent can reach to one approved vault item at a time, but only aac run keeps the secret out of the…
-
SoduWork Turns AI Agents Into a Team That Needs Your Sign-Off
SoduWork moves the finish line of agent work from the agent's answer to written acceptance criteria, a reviewer on a different model and a…
-
landmines Turns Your Linter Into a Lie Detector for AI-Written Tests
A green CI run from a coding agent proves little when the agent is free to mock the code under test or retry the flake, and landmines moves…
-
Java4Shell Was Found With AI in an Hour. Here's Your 24-Hour Patch Plan
When an AI-assisted researcher can find a JVM memory bug in an hour and weaponize it in an afternoon for under $200, the gap between…
-
Codex Sandboxes the Commands Its Model Runs. Your MCP Servers Run Outside the Fence.
Codex fences the commands its model runs but launches stdio MCP servers with your full user permissions, so wrapping each server in codex…
04 / Coverage map
Topics we track
Claude Code 125 Anthropic 53 OpenAI 44 GitHub Copilot 26 Codex 25 Agent Skills 19 Hugging Face 17 MCP 17 LangChain 15 Cloudflare 14 DeepSeek Harness 14 Model Context Protocol 13 Kimi K3 11 METR 11 Anthropic Frontier Red Team 8 GPT-5.6 Sol 8 Claude Code auto mode 7 GPT-6 Astra 7 MCP 2026-07-28 7 Claude Fable 5.1 6 Codex CLI 6 Cursor 6 GPT-5.6-Cyber 6 LangChain Deep Agents 6