Local video editor for coding agents: 42 FFmpeg tools with a machine-readable contract and an MCP server, for Claude Code, Cursor and Codex. Cut, captions, reframe, loudness, platform exports. No cloud, no API keys.
Python agent-skillsai-agentsaudio-processingcaptionsclaude-codecodexcoding-agentcursorffmpegloudness-normalizationmcpmcp-servermodel-context-protocolpythonsubtitlestranscodingvideovideo-editingvideo-processingwhisper
Catch CHANGELOG drift automatically; document the pitfall (#90)
Shogo Kajiyama81fd69e9eda8
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.ymlindex 490e4f5..1e0899a 100644--- a/.github/workflows/ci.yml+++ b/.github/workflows/ci.yml@@ -18,6 +18,12 @@ jobs:runs-on: ${{ matrix.os }}steps:- uses: actions/checkout@v4+ with:+ # Full history (not the default shallow depth-1 clone): test_contract.py's+ # test_changelog_mentions_every_closed_issue_since_last_tag walks git log back to the+ # latest tag to check CHANGELOG.md against "Closes #N" in commit bodies -- a shallow+ # clone would make that check silently a no-op (no tag reachable to diff against).+ fetch-depth: 0- uses: actions/setup-python@v5with:python-version: "3.9"diff --git a/CHANGELOG.md b/CHANGELOG.mdindex b784dc1..09f9caa 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -50,7 +50,8 @@ field addition here is additive.consistently instead of silently defaulting to track 0 (or, for `fit.py`, to ffmpeg's ownimplicit "best stream" heuristic, which for audio favours channel count over track order).`join.py`/`multicam.py` are out of scope — they combine separate input files, a different- problem shape.+ problem shape. Closes [#62](https://github.com/kajisho5/ffmpeg-skill/issues/62) (the same gap+ `caption.py`/`audio.py` already closed in [#55](https://github.com/kajisho5/ffmpeg-skill/issues/55)).- **`doctor --json` gains a `fonts` field**, informational like `gpu_encoders`: drawtext's defaultfont (`caption.py --animate`/`--karaoke`, `graphics.py`) can silently substitute a differentfamily when the requested one isn't installed — a drawtext exit code can't detect thisdiff --git a/references/process-pitfalls.md b/references/process-pitfalls.mdindex a7b2d83..e01479a 100644--- a/references/process-pitfalls.md+++ b/references/process-pitfalls.md@@ -63,3 +63,23 @@ on the *same* platform in a different way: stop redesigning the fixture. Eitherthe strict assertion to the platform where it's provably correct (keeping a weaker,platform-general check — output exists, has the right duration — everywhere), or escalatebefore spending a third CI cycle on it.++## A fix merged after CHANGELOG.md's current-version section was drafted can silently miss it++`CHANGELOG.md`'s `## 0.12.0` section was written once, covering everything merged up to+that point. Two fixes that closed real issues after that point (#62's `--audio-stream`+extension via PR #72, #77's dry-run-dims fix via PR #88) landed with no further nudge to+go back and add a bullet — #62's fix actually got a bullet (its content is genuinely+described) but the `Closes #62` link was left off, and #77 was missed outright until a+direct question ("shouldn't this bump the version?") prompted a manual check. Neither was+caught by CI, because nothing checked CHANGELOG.md against what had actually been closed.++Caught by hand both times, then closed properly with `tests/test_contract.py`'s+`test_changelog_mentions_every_closed_issue_since_last_tag`, which walks `git log` back to+the latest release tag, extracts every `Closes #N.` from a commit body, and fails if that+issue number doesn't appear anywhere in `CHANGELOG.md`. This needs real history (`ci.yml`'s+`actions/checkout` step now passes `fetch-depth: 0` for exactly this reason — the default+shallow clone leaves no tag reachable to diff against, which would make the test silently+skip itself in CI, not fail). If this test ever needs to skip a genuinely changelog-less+closed issue (a pure process note, a duplicate, a revert of an unreleased change), name the+exemption in the test itself with a reason — don't just widen the regex or drop the check.diff --git a/tests/test_contract.py b/tests/test_contract.pyindex 750e7fa..ecc726a 100644--- a/tests/test_contract.py+++ b/tests/test_contract.py@@ -359,6 +359,25 @@ class ContractTests(unittest.TestCase):self.assertEqual(meta["video"]["height"], 0)self.assertEqual(meta["video"]["fps"], 0.0)+ def test_changelog_mentions_every_closed_issue_since_last_tag(self):+ """A merged fix can land after CHANGELOG.md's current-version section was already+ written (this happened for real: #77's fix, PR #88, merged after the 0.12.0 section was+ drafted and initially missed it -- caught by hand, not by a test, and fixed in a+ follow-up commit). Every "Closes #N." in a commit body since the last release tag should+ show up somewhere in CHANGELOG.md; if it doesn't, either the changelog needs an entry or+ the issue was closed without one on purpose (rare -- e.g. a pure process/doc note) and+ this test's exemption set below should say why. Needs real git history: skips itself on+ a shallow clone (see ci.yml's fetch-depth: 0) where no tag is reachable at all."""+ tags = sh("git", "tag", "--list", "v*", "--sort=-creatordate", cwd=ROOT, check=False).stdout.split()+ if not tags:+ self.skipTest("no reachable release tag (shallow clone?) -- nothing to diff against")+ last_tag = tags[0]+ log = sh("git", "log", f"{last_tag}..HEAD", "--format=%B----COMMIT----", cwd=ROOT, check=False).stdout+ closed = sorted(set(int(n) for n in re.findall(r"(?im)^closes\s+#(\d+)\.?\s*$", log)))+ changelog = (ROOT / "CHANGELOG.md").read_text(encoding="utf-8")+ missing = [n for n in closed if f"#{n}" not in changelog]+ self.assertEqual(missing, [], f"issue(s) closed since {last_tag} but not mentioned in CHANGELOG.md: {missing}")+def test_dry_run_metadata(self):for t in self.contract["tools"]:has_flag = "dry_run" in t["input_schema"]["properties"]