Repository navigation
Make AGENTS.md canonical and reach every runtime (stack 3/5) - #44
Closed
martient wants to merge 1 commit into
Conversation
martient
added this pull request to stack #48
September 20, 2026 17:40
AGENTS.md and CLAUDE.md were near-identical files kept in sync by hand; the diff was two lines and had already started drifting (the trailer ban landed in one of them only). AGENTS.md is now the single source and CLAUDE.md is an @AGENTS.md import plus the one genuinely Claude-specific line, which is the pattern the ecosystem settled on. Two runtimes read neither file, so they now get thin pointers: .github/copilot-instructions.md and GEMINI.md. Skills are exposed at .agents/skills, the cross-tool discovery path that Codex, Claude Code and Copilot CLI all resolve, via a symlink to the canonical copy under plugins/committy/skills -- no duplication. A test guards the symlink, because a checkout that materialises it as a plain file would break discovery silently. Skill frontmatter gains the spec's optional fields: license, allowed-tools (so security-conscious runtimes can pre-authorise exactly what each skill needs) and metadata carrying a skill version independent of the CLI version. committy-release referenced siblings with Codex's `$skill` syntax inside a body shared by both runtimes; they are now named plainly and each runtime resolves them its own way. The COMMITTY_NONINTERACTIVE docs test now accepts a file that defers to AGENTS.md rather than restating it, which is what the split requires.
martient
force-pushed
the
claude/ai-agents-capabilities-report-fgpp3p-cross-tool
branch
from
September 20, 2026 18:03
f8793fe to
30ca219
Compare
Owner
Author
|
Superseded by #49, which carries these commits unchanged. Closing — see #42 for why the stack was collapsed. Generated by Claude Code |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked on #43. Resolves P1-8, P1-9, P1-10, P1-11 and P2-14.
One instruction file, not two
AGENTS.mdandCLAUDE.mdwere near-identical files kept in sync by hand — a two-line diff that had already started drifting: the AI-attribution-trailer ban landed inCLAUDE.mdonly.AGENTS.mdis now the single source;CLAUDE.mdis an@AGENTS.mdimport plus the one genuinely Claude-specific line (slash-command invocation), which is the pattern the ecosystem settled on.Two widely-deployed runtimes read neither file, so they get thin pointers:
.github/copilot-instructions.mdandGEMINI.md.Cross-tool skill discovery
.agents/skillsis the path Codex CLI, the Codex app, Claude Code and Copilot CLI all resolve. It now exists as a symlink to the canonicalplugins/committy/skills, so there is no second copy to drift.core.symlinks=false— silently breaks discovery for every runtime reading that path.test_skills_are_discoverable_at_the_cross_tool_pathfails loudly in that case rather than letting it pass unnoticed. If you'd rather not depend on symlinks at all, the alternative is moving the canonical copy to.agents/skillsand pointing both plugin manifests there; I didn't take it because I can't verify from here whether Claude'splugin.jsonaccepts askillspath outside the plugin directory.Skill frontmatter
All four skills carried only
nameanddescription. They now add the spec's optional fields:license: Apache-2.0allowed-tools— so security-conscious runtimes pre-authorise exactly what each skill needs (committy-enforceis the only one grantedEdit/Write, since it installs hooks and CI)metadata— a skill version independent of the CLI version, plus the minimum Committy it expectsCodex syntax in a shared body
committy-release/SKILL.mdreferenced siblings as$committy-commit. That body is shared by both runtimes, and Claude Code invokes them as/committy:committy-commit. Siblings are now named plainly and each runtime resolves them its own way.Test contract change
test_agent_docs_mention_the_non_interactive_env_varpreviously required every agent-facing file to containCOMMITTY_NONINTERACTIVEverbatim. With one canonical file that is wrong, so it now accepts a file that either states the rule or defers toAGENTS.md— and covers the two new pointer files too.Verification
375 passed,
fmtclean,clippy -D warningsclean. The pre-existingtag_testsfailure noted in #42 is unchanged.