-
Gitea user account used by Claude Code with Gitea MCP to manage repositories: read/write issues and PRs, comment on changes, and perform merge/review actions with appropriate permissions.
- Joined on
2026-06-20
Resolved on branch docs/simplification-audit, not yet merged:
- The merge:
skill-auditandagent-auditare merged intofactory-audit(commit 620f20b, ADR-0025), and the call sites in…
Opened #133 implementing this — moved and renamed all 12 files to lowercase-kebab-case under references/ across the four skills, fixed the cross-skill citation and every in-skill link (including…
Opened #132 implementing this — pinned gitea-mcp@v1.7.0, portable bare go command, GITEA_ACCESS_TOKEN/GITEA_HOST required via ${VAR} expansion with no inline defaults or hardcoded…
Closing without action — reviewed during grill-with-docs triage of the small-issue batch. Pre-commit installation and successful hook execution are already exercised incidentally across the…
Full review done, findings fixed and pushed (4f4b55b..0f2bb24). Six commits, split by concern.
**Five blockers, all of the same shape — a gate or check that reported success without measuring…
user-invocable: false as a hand-invocation marker, which is backwards
Implementation note
Confirmed via agent-definition.md's field table that disable-model-invocation and user-invocable are independent: the first controls whether the runtime auto-selects…
Implementation note
Two approaches from the suggested fix were tried and rejected before implementing, worth recording so they aren't re-proposed blind:
- Literal filename cross-check…
metadata.version has no owning rule
Implementation note (follow-up — git-plugin portion)
Completed the remaining slice: git-branches, git-history, git-remotes, git-submodules, git-workflow, git-worktrees, pc-author…
rtk git vs bare git is inconsistent across skills
Implementation note
Rule decided in session, empirically validated against git-commits before the corpus-wide pass (every existing bare git there turned out illustrative, never an…
metadata.version has no owning rule
Implementation note
Decision (discussed and confirmed in session, recorded in docs/adr/0022-skill-metadata-version-is-mandatory.md): metadata.version moves from a per-plugin, undocumented…
Already resolved, closing without further changes. Verified against main (post 598a7c3 / #129): the ownership split is git-commits owns cherry-pick, git-history owns revert — not the "one…