Fetching from the wire…
Tools2026-09-07 · source-backed
PR #43442 closes two gaps (GitHub). Concurrent parent compaction could remove evidence between checkpoint selection and prompt construction, so both now read the same parent history snapshot. Separately, a finished allow decision is cancelled if the session's user-message revision or root authorization version changed during review, meaning an approval granted against one set of instructions can't be spent after the user supplied different ones. That's the time-of-check-to-time-of-use bug applied to human approval, and I hadn't thought about it in those terms before.
Each link below shares sources, entities, or timing with this story.
Rust, created May 14, at 2,643 stars (GitHub). Every run produces checkpoints linking a commit to the session that made it, including prompts, tool calls and reasoning. It runs Claude Code, Codex, its own agent and anything from the ACP registry side by side against one codeba...
GitHub published four advisories against omnigent-ai/omnigent v0.1.0, the meta-harness that runs Claude Code, Codex and Pi under policy and sandboxing. GHSA-jrrm-9hc7-2v3h at CVSS 9.0 lets any user with session edit rights overwrite a shared template agent via PUT /sessions/{i...
5.8k stars, Rust, macOS, ~40 MB versus ~67 MB upstream, keeping full Lua customization. Command-failure recovery with suggested fixes applied via Cmd + Shift + E, natural-language-to-command via # <description>, preconfigured integration for Claude Code, Codex, Gemini CLI, and...
affaan-m/ECC (36.3k forks, MIT) bundles 67 agents, 284 skills, 94 legacy command shims, and "instincts", patterns learned from prior sessions with confidence scores that auto-recall when relevant, plus a .ecc/memory/ markdown vault that's explicitly cross-harness, so context s...
EchoVault solves the "agent amnesia" problem — coding agents forgetting everything between sessions. It runs as an MCP server giving agents three tools: memory_context (load prior decisions), memory_search (find specific memories), and memory_save (persist learnings). Storage...
The result-interception hook (#41202) is a place to redact secrets or strip injected instructions out of untrusted MCP output before it enters context, which is the missing seam in every MCP setup I've built. Also: a configurable grace period for discovering tools from optiona...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.