feat: add crawl archive skills#83040
Conversation
|
Codex review: needs real behavior proof before merge. Summary Reproducibility: not applicable. This PR adds optional bundled skill definitions rather than fixing a reported current-main bug. The reviewable behavior is source-verifiable metadata plus installer provenance, not a user reproduction path. Real behavior proof Next step before merge Security Review detailsBest possible solution: Maintainers should decide whether these ClawHub-published crawler skills warrant bundled promotion; if yes, keep the canonical OpenClaw module metadata and add a disposable install smoke before landing. Do we have a high-confidence way to reproduce the issue? Not applicable: this PR adds optional bundled skill definitions rather than fixing a reported current-main bug. The reviewable behavior is source-verifiable metadata plus installer provenance, not a user reproduction path. Is this the best way to solve the issue? Unclear until maintainer approval: the current patch is internally coherent after the module-path fix, but VISION.md makes bundled promotion a human product/security decision over the ClawHub-only path. Acceptance criteria:
What I checked:
Likely related people:
Remaining risk / open question:
Codex review notes: model gpt-5.5, reasoning high; reviewed against 51e93669cb4f. |
51acf67 to
77e9b4e
Compare
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 77e9b4e441
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".
| - slacrawl | ||
| install: | ||
| - kind: go | ||
| module: github.com/vincentkoc/slacrawl/cmd/slacrawl@latest |
There was a problem hiding this comment.
Point installer modules at OpenClaw-owned crawl repos
The new crawl skills advertise OpenClaw ownership via metadata.openclaw.homepage, but the install module here (and the matching lines in graincrawl/notcrawl) still points to github.com/vincentkoc/.... Because the installer uses this module path to fetch binaries, users can end up installing from a different owner than the declared homepage, which breaks provenance and can pull stale/divergent code if those repos drift. Update these module values to the OpenClaw module paths to keep install source and ownership metadata consistent.
Useful? React with 👍 / 👎.
|
Landed via temp rebase onto Behavior addressed: bundled crawl archive skills added for Discrawl, Gitcrawl, Graincrawl, Notcrawl, and Slacrawl; install recipes corrected to declared Go module paths before merge.
Source PR head: 77e9b4e |
Summary
Verification
Behavior addressed: stock OpenClaw crawler skills now have canonical repo-local definitions, OpenClaw repo/module metadata, and root ClawHub slugs.
Real environment tested: local OpenClaw checkout and live ClawHub registry.
Exact steps or command run after this patch: frontmatter YAML parse, git diff --check, node scripts/run-vitest.mjs src/agents/skills/frontmatter.test.ts, clawhub inspect for all five root slugs.
Evidence after fix: each root slug resolves under openclaw; Slacrawl, Graincrawl, and Notcrawl latest metadata versions point at OpenClaw repos/modules.
Observed result after fix: root installs/searches resolve to OpenClaw-owned stock skill entries.
What was not tested: installing each skill through a fresh OpenClaw runtime.