Daily Edition Sources +2

Codex Moves Skill Instructions Into Model Metadata

A compact Codex patch makes skills usage instructions depend on model metadata instead of hardcoded model-name matching.

Diagram Punk poster showing Codex model metadata feeding core and extension skill renderers while hardcoded model matching is crossed out.
Diagram Punkskill rules moved into model metadata.
repo openai/codex evidence
2 source signals 1 repo commit 6b5f574
Evidence: commit 6b5f574 / June 29, 2026 / Daily Edition
Open Edition Evidence below

Signal: Codex was yesterday's lead. Commit 6b5f574, merged on June 29, is a narrower follow-up worth tracking because it moves skill behavior into model metadata.

Public source: PR #29740 adds a false-by-default include_skills_usage_instructions model metadata field, enables it for bundled gpt-5.5 metadata, consumes it in core and extension skill rendering, and removes the old hardcoded legacy-model path.

Implication: skill guidance is becoming a model capability flag. That is a cleaner integration contract for agent builders than asking code paths to infer behavior from model names.

Caveat: the evidence does not show a public release or a general OpenAI API rule. It is a Codex repository change with tests across app-server, core, extensions, and model metadata.

What to watch: more agent UI behavior may move into explicit model metadata. This stays compact because it continues yesterday's Codex trust/runtime beat and is narrower than Tau's new public onboarding surface.

Evidence Trail

Receipts below the story

The article above is the public narrative. This section keeps the source trail and limits on the same page.

Edition
DateJune 29, 2026
LaneDaily Edition
Confidence68%
Sources2
Reposopenai/codex

Primary Evidence

  • Codex model metadata skill-instruction patch: https://github.com/openai/codex/commit/6b5f5743b3169ef463155241da2bab6888a3cbe4
    • Shows: adds include_skills_usage_instructions model metadata, enables it for bundled gpt-5.5, consumes it in core and extension skill rendering, and removes hardcoded legacy-model matching.
  • Codex PR #29740: https://github.com/openai/codex/pull/29740
    • Shows: public PR receipt for the same patch and test surface.
  • Prior Codex runtime-policy article: /articles/2026-06-28-daily-codex-makes-plugin-trust-runtime-policy/
    • Shows: continuity check explaining why this was a compact brief rather than the lead.

Evidence Limits

  • The evidence proves a Codex repository change and tests. It does not prove a public release, API-wide behavior, or a general rule for all model metadata.
Letters & Corrections

Send a note to the desk

Corrections, missing context, or a follow-up lead.