697d4ce974
Adds an `agents/openai.yaml` beside each `SKILL.md` so the skills work in Codex as well as Claude Code, without generated copies: - `interface.display_name` + `interface.short_description` for the Codex skill picker, hand-written for all 39 skills. - `policy.allow_implicit_invocation: false` on the 22 user-invoked skills — the Codex analog of `disable-model-invocation: true`, so Codex excludes them from implicit invocation while explicit `$skill` still works. - Document the dual-harness invocation model in `.agents/invocation.md`, `CLAUDE.md`, and the promoted-bucket READMEs. - Add `AGENTS.md` as a symlink to `CLAUDE.md` so Codex reads the same instructions; note Codex as a `link-skills.sh` install target. Slimmed-down rework of the approach prototyped in #522: keeps the essential cross-harness metadata, drops the Ruby validator, the runtime-detector test, per-skill `default_prompt`s, and the unrelated promotion changes. Co-authored-by: gabimoncha <gabimoncha@users.noreply.github.com> Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2.3 KiB
Model-invoked vs user-invoked
Every SKILL.md in this repo is a skill. The one axis that splits them is invocation — who can reach it:
- User-invoked — reachable only by the human typing its name. Set
disable-model-invocation: truein the frontmatter (Claude Code) andpolicy.allow_implicit_invocation: falseinagents/openai.yaml(Codex). Thedescriptionis human-facing: a one-line summary read by a person browsing slash-commands. Strip trigger lists ("Use when the user says…"). - Model-invoked — reachable by model or user. The default: omit
disable-model-invocationand thepolicyblock fromagents/openai.yaml. Thedescriptionis model-facing and keeps rich trigger phrasing ("Use when the user wants…, mentions…, asks for…") so auto-invocation fires. The test for whether a skill should stay model-invoked: could the model usefully reach for this autonomously? (Reuse is the reason to extract a skill, not the test.)
Each harness excludes a user-invoked skill from the model's reach in its own way, so nothing but the human can fire it — no other skill can. A user-invoked skill may invoke model-invoked skills, but it can never reach another user-invoked skill.
Every skill also carries an agents/openai.yaml beside its SKILL.md. It holds Codex UI metadata — interface.display_name and interface.short_description for the skill picker — and, for user-invoked skills, the policy.allow_implicit_invocation: false that pairs with disable-model-invocation. Keep the two in sync: a skill is user-invoked in both harnesses or neither.
Bucket README.mds and the top-level README.md group entries into User-invoked and Model-invoked.
Dependencies between them
Dependencies are expressed as /skill-style prose invocation ("Run the /grilling skill"), not deep ../other-skill/FILE.md cross-references. Shared reference docs live inside the skill that owns them; other skills reach that material by invoking the skill, not by linking across folders.
Passive vs active domain work
Merely reading CONTEXT.md for vocabulary is a one-line prose pointer, not the domain-modeling skill. Only the active build/sharpen discipline (challenge terms, edge-case scenarios, write ADRs, update CONTEXT.md inline) is domain-modeling.