spec-grill
spec-grill
Author spec/capabilities.md, the middle layer between spec/charter.md and day-to-day execution work. The file is optional: create it only when a keep condition holds, amend it if it already exists. It is not a file generated from a directory listing, and not the second step of a required pipeline.
Execution contract
Intent router
Report by default. Read the request and take the safest reading of it; the user never has to memorize arguments or mode names. Write to spec/capabilities.md only on clear edit intent — "write it", "add the missing capability", "문서 적을 건 적고" — or after the user confirms a proposed edit. Everything else — diagnosis, candidate discovery, single-capability review, and an audit that judges existing contracts for stale, overlapping, weak, or unsupported predicates — ends in a report, and when intent is unclear, prefer report-only. Capability slugs are lowercase singular handles; the nuance belongs in Goal/Scope prose, not in the slug.
Helper scripts
Resolve helper scripts from the installed spec-grill skill directory, not from the target repo. In a source checkout, that means the local scripts/ directory beside this SKILL.md. Always pass the target repo explicitly (--repo-root <target-repo>) so helpers do not inspect the skill directory by accident.
On a brownfield repo with no spec/capabilities.md, or when candidate evidence is requested, run node <skill-dir>/scripts/extract-signals.js --repo-root <target-repo> --json first. The script reports raw capability evidence. It never writes spec/capabilities.md; admission, merging, splitting, and naming belong to this skill.
Use-or-delete checkpoint 2026-12-20: see docs/status.md.