Graphiti memory integration (ingest + recall)

[yulongbai-nov]

Summary

Integrates Graphiti-backed memory into Codex CLI:

• async turn ingestion to Graphiti (POST /messages) for session/workspace scopes
• optional per-turn recall (POST /search) injected as a <graphiti_memory> system message

Safety & performance

• opt-in ([graphiti].enabled), explicit consent, and trusted-project gating
• fail-open + timeouts; ingestion off the hot path (bounded queue + retry)

CLI

• codex graphiti status
• codex graphiti test-connection
• codex graphiti promote
• codex graphiti purge

Docs & specs

• .specs/graphiti-memory-integration/{design,requirements,tasks}.md
• docs/demos/graphiti-memory-integration/README.md
• docs/config.md + docs/example-config.md

Tests

• codex-rs/core/tests/suite/graphiti_memory.rs
• unit test for git source_description formatting

Notes: relaxed the timing assertion in codex-rs/core/tests/suite/tool_parallelism.rs to reduce flakiness under loaded test runners.

[github-actions[bot]]

Thank you for your submission, we really appreciate it. Like many open-source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution. You can sign the CLA by just posting a Pull Request Comment same as the below format.

---

I have read the CLA Document and I hereby sign the CLA

---

_{You can retrigger this bot by commenting recheck in this Pull Request. Posted by the CLA Assistant Lite bot.}

[etraut-openai]

I'm not sure if you intended to create a PR in this repo.

We've updated our contribution guidelines to clarify that we're currently accepting contributions for bugs and security fixes, but we're not generally accepting new features at this time. If you would like to propose a new feature, please file or upvote an enhancement request in the issue tracker. We will generally prioritize new features based on community feedback.