This release adds 3 notable features for engineering teams evaluating rollout.
Published 1mo
AI Agents & Assistants
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
agent-harness
agent-isolation
agentic
ai-agents
ai-coding
ai-tools
+14 more
aider
claude-code
code-provenance
codex
coding-assistant
cursor
developer-tools
gemini-cli
git
git-worktree
llm-agents
python
vcs
worktree
Summary
AI summaryCross-agent recall enables Claude and Codex to see each other's decisions via transcript memory notes.
Full changelog
Brings Codex CLI to feature parity with Claude Code in the ait transcript memory pipeline.
Added
ait adapter setup codexinstalls a native Codex hook bridge under.ait/adapters/codex/codex_hook.pyand merges the matching hook config into.codex/hooks.json.- Each Codex session becomes one ait attempt with the rollout jsonl persisted under
.ait/transcripts/<attempt-id>.jsonl— same as Claude Code. - Cross-agent recall now flows in both directions: Claude can see what Codex decided last week, and vice versa, via the existing transcript-summary memory notes.
Fixed
- Hook bridges no longer send empty-string
modelin theattempt_startedpayload. The protocol validator rejects empty strings, which previously surfaced as "daemon closed the connection" warnings on agents that omitmodel.
Setup
ait adapter setup codex
ait adapter doctor codex
codex # session is now recorded by ait
Install: pipx install ait-vcs or npm install -g ait-vcs.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About Ait
All releases →Related context
Related tools
Earlier breaking changes
- v0.55.52 Keep the built-in claude-code reviewer pinned to the local claude -p CLI even when repository policy defines a conflicting command override.
Beta — feedback welcome: [email protected]