feat(cli-registry): nine more tools + version capture, recovered from a worktree - #2725
Conversation
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e09a38b143
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
|
Both P2s addressed in |
663eb82 to
b4afcdb
Compare
… a worktree Found uncommitted in the pmoves-cli-registry worktree during a worktree sweep. #2599 (feat/cli-tools-registry) had merged; this work was done afterwards and never committed, so it existed only in that working tree. WHAT IT ADDS host_clis tailscale, opencode, node, npm project_devtools jest, eslint pinokio_ecosystem pinokio, pterm, gepeto install_tools.py _cli_version() - best-effort version capture: runs the manifest's own check command, keeps line 1, and swallows OSError/ValueError/TimeoutExpired so a broken or missing CLI degrades to "no version" rather than taking the whole --check run down. pterm and gepeto matter beyond the count: they are the Pinokio-ecosystem tools that were being run from a downloaded install rather than the fork, which is the conflict this registry exists to make visible. The tailscale entry carries its own warning in the purpose string - raw `tailscale status` prints fleet IPs, so the manifest points at the make target instead. Worth keeping as written rather than trimming to a one-liner. CONFLICT RESOLUTION main had since added a `skills:` entry at the same position in host_clis, so the 3-way apply conflicted. Both sides are additive entries in the same YAML map, so both were kept rather than either being taken. Verified afterwards: host_clis holds skills AND tailscale/opencode/node/npm, and no conflict markers survive. VERIFIED `install_tools.py --check` runs against the merged manifest and correctly resolves tailscale, node and npm on this host while reporting opencode and skills as optional-missing. An earlier check of mine looked for these keys at the top level and reported them all MISSING - the file is categorised (host_clis / project_devtools / pinokio_ecosystem), so that was my lookup being wrong, not the merge. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
… their workspace --check now passes report_versions (default on, --no-report-versions opts out): the parser never exposed a flag and main() never passed the argument, so _cli_version() was dead code and the advertised version capture printed nothing. jest/eslint run_via now point at the owning workspace (npm --prefix pmoves/ui ...): run from the repo root there is no package.json, so `npm test` fails outright and bare `npx jest`/`npx eslint` resolve unrelated versions instead of the ones the ui lockfile pins.
b4afcdb to
cdac816
Compare
Found uncommitted in the
pmoves-cli-registryworktree during a worktree sweep. #2599 had already merged; this work was done afterwards and never committed, so it existed only in that working tree.What it adds
host_clistailscale,opencode,node,npmproject_devtoolsjest,eslintpinokio_ecosystempinokio,pterm,gepetoPlus
install_tools.py::_cli_version()— best-effort version capture that runs the manifest's own check command and swallowsOSError/ValueError/TimeoutExpired, so a broken or missing CLI degrades to "no version" rather than taking the whole--checkrun down.ptermandgepetomatter beyond the count — they're the Pinokio-ecosystem tools that were being run from a downloaded install rather than the fork. That conflict is exactly what this registry exists to make visible.The
tailscaleentry carries its own warning in the purpose string: rawtailscale statusprints fleet IPs, so the manifest points at the make target instead. Kept as written rather than trimmed.Conflict resolution
mainhad since added askills:entry at the same position inhost_clis, so the 3-way apply conflicted. Both sides are additive entries in the same map, so both were kept rather than either taken.Verified after:
host_clisholdsskillsandtailscale/opencode/node/npm, with zero surviving conflict markers.Verified
install_tools.py --checkruns against the merged manifest and correctly resolvestailscale,nodeandnpmon this host, while reportingopencodeandskillsas optional-missing.An earlier check of mine looked for these keys at the top level and reported them all MISSING. The file is categorised (
host_clis/project_devtools/pinokio_ecosystem) — that was my lookup being wrong, not the merge. Worth stating because "the tools are missing" and "I looked in the wrong place" produce identical output.🤖 Generated with Claude Code