3 Commits

Author	SHA1	Message	Date
Oleks	832b07b248	tests: make the worktree-discipline remote-resolution test hermetic ... kotkan/claude-plugin-inference-arbitrage#30 reported "FAIL repo = kotkan/worktree-discipline (want kotkan/claude-plugin-worktree-discipline)". Investigation found no stale reference in this repo's source or fixtures — the expected value already matches the live repo (verified: only kotkan/claude-plugin-worktree-discipline exists on Gitea; the pre-rename bare-name slug is a redirect only). The failure came entirely from the live git remote of the sibling ../worktree-discipline checkout on disk, which is outside this repo and outside version control, so it can drift independently (e.g. an old clone never repointed after the 2026-07-22 kotkan rename). That also meant the assertion silently never ran in CI, since the bare CI container has no sibling checkout and the test just printed "skip". Replace the dependency on that external checkout with a throwaway git repo created inside the test's own tmpdir, with its remote set explicitly to the current repo slug. Same code path (resolve_repo/parse_remote in bin/filing-plan) is exercised deterministically, and the assertion now always runs instead of skipping outside a workspace with that sibling checkout present. Verified the new test both passes against the correct remote and correctly reproduces the exact original failure signature when pointed at the stale pre-rename slug.	2026-07-30 23:27:11 +03:00
Oleks	b77fbdddbb	Fix marker-wrapper loss and add backward-compat match (kotkan/claude-plugin-inference-arbitrage#23) ... Root cause: bin/filing-plan and references/issue-template.md always render the ia-candidate marker HTML-comment-wrapped. A direct issue_write probe confirmed the Gitea MCP write path preserves a literal <!-- --> unchanged, so the loss on oleks/claude-plugin-cluster#56 and #57 happened on the Agent(anxious:issuer-agent) relay hop: an HTML comment is invisible when its surrounding text is read as markdown, so an agent relaying "what the body should say" instead of pasting it verbatim can silently drop it. Two fixes: - bin/filing-plan: find_existing()'s marker_re() now also matches the bare, unwrapped "ia-candidate: <id>" line (end-of-line anchored, not \b, since candidate ids contain "-" and a word-boundary check would false-match a longer id sharing a prefix) so already-filed issues like cluster#56/#57 are still recognized and not duplicate-filed. - skills/offload-audit/SKILL.md: the issuer delegation prompt now fences the rendered body in its own code block and explicitly warns that the leading marker is an invisible-by-design HTML comment, not stray syntax to clean up, instead of inlining it as bare markdown the relaying agent could read right through. references/issue-template.md documents both, and tests/filing.test.sh adds coverage for the bare-marker match plus a same-prefix-different-id non-match case. Bumps plugin.json to 0.7.1.	2026-07-30 19:24:13 +03:00
Oleks	5bdc984674	Fix markdown lint in the Phase 5 docs	2026-07-29 17:50:20 +03:00