Catalog
fairoosansar-coder/review-pull-request

fairoosansar-coder

review-pull-request

Review pull requests for opentelemetry.io: CI check semantics, CLA and approval-label workflow, refcache handling, locale rules, and content quality. Use when reviewing a PR or debugging a CI failure in open-telemetry/opentelemetry.io.

v1.0LATEST
NewUpdated Oct 29, 2026

Review Pull Request

Review workflow for pull requests in open-telemetry/opentelemetry.io. The contributing guide and the per-check decoder in pr-checks.md are the authoritative sources — when this skill drifts from them, trust them.

For blog-specific rules (gh-url-hash, author front matter, publish-date gating), defer to the sibling review-blog-post skill. For label drafting guidance, defer to draft-issue.

Arguments {#arguments}

  • If $ARGUMENTS is empty, ask for a PR number or URL.
  • If $ARGUMENTS is a GitHub URL containing /pull/, extract the numeric PR number after /pull/.
  • If $ARGUMENTS starts with #, strip the # and use the digits.
  • If $ARGUMENTS is a bare number, use it.
  • Otherwise, stop and ask for a valid PR number or URL.

When to use

  • Reviewing a PR in open-telemetry/opentelemetry.io.
  • Debugging a CI check failure on a PR.
  • Preparing your own PR for submission.

Workflow

1. Setup

Pull metadata, diff, and checks; classify changed files:

gh pr view <N> --json title,body,files,reviews,labels,author,isDraft,headRepositoryOwner
gh pr diff <N>
gh pr checks <N>

Group files into content/en/blog/**, content/en/docs/**, content/<lang>/**, data/registry/**, .github/**, scripts/**, or config — classification drives which CI checks matter and which rules apply.

2. Walk CI checks

For each failing check, match <workflow-name> / <job-name> against pr-checks.md — every check has a section describing what it validates and the local fix command. Caveats:

  • Link checking is sharded (en / locales-A-to-M / locales-N-to-Z); a single failing shard does not necessarily block merge — read the failure.
  • Fork PRs can hit token-scope limits that look like check failures but are permissions artifacts. Read the log before concluding.
  • Netlify Deploy Preview failures: open Details for the build log before reasoning about them.

3. Verify process rules

Pre-merge approval

Content origin

Branch state

4. Review content

For docs PRs (content/en/docs/**):

  • Front matter. Valid YAML; appropriate title, linkTitle, weight, description; Hugo-specific fields intact.
  • Terminology. "OpenTelemetry" one word; "OTel" only after first full mention; signal names lowercase (traces, metrics, logs); component names cased (SDK, API, Collector); proper nouns cased. Enforced by textlint via .textlintrc.yml.
  • Link references. Prefer collapsed form [text][] over shortcut [text]; enforced by scripts/_md-rules/no-shortcut-ref-link/.
  • Markdown extensions. GitHub alerts and Obsidian callouts are OK.
  • Internal links. Use Hugo ref / relref or paths starting with /docs/...; never full https://opentelemetry.io URLs.
  • Code blocks carry a language tag; images carry meaningful alt text.

For blog PRs (content/en/blog/**), defer to the review-blog-post skill.

5. Final pass and output

Walk this checklist before writing the review:

CI and process

  • Easy CLA green (or author has a fix path).
  • Netlify preview builds.
  • Each failing check-* assessed against pr-checks.md#checks.
  • Linked issue is triage:accepted (or this is an auto/hotfix PR).
  • Does not span locales with semantic changes — or uses # patched for editorial cross-locale edits.
  • First-time-contributor AI checklist in the PR description is filled in and looks human-written.
  • No unrelated changes bundled.

Labels

  • Auto-applied labels look correct (sig/lang/blog/registry/i18n); none added by hand.
  • ready-to-be-merged / missing:* not touched manually.
  • sig-approval-missing added if docs approval landed without SIG approval on a co-owned PR.

Content

  • Front matter valid; terminology consistent; code blocks tagged; images have alt text; internal links use paths or Hugo refs (not opentelemetry.io URLs); no shortcut-form reference links.

Refcache and links

  • refcache.json updates (if any) committed in the PR.
  • No hand-edits to refcache.json.
  • Unreachable-but-valid URLs use ?link-check=no (see Refcache).

Then structure the review as:

  • CI Status Summary — one line per check (pass/fail/skip); call out fork-PR permissions artifacts separately from real failures.
  • Required Changes (Blocking) — issues that must be fixed before merge. Cite a file or check name for each.
  • Suggested Improvements (Non-blocking) — terminology, cross-link opportunities, phrasing.
  • Positive Feedback — short but present.

Refcache {#refcache}

static/refcache.json is a 1MB+ cache of external-link status codes. npm run check:links updates it as a side effect — authors commit the updated file themselves (pr-checks.md#build-and-check-links). The Links / REFCACHE updates? job fails if the on-branch cache is stale relative to what the link check produced.

Do not hand-edit refcache.json. If a URL returns a non-200 for server reasons (blocked bot, LinkedIn 999, …), append ?link-check=no (or &link-check=no) to the URL — pr-checks.md#handling-valid-external-links. Maintainers can validate 4xx entries via ./scripts/double-check-refcache-4XX.mjs.

For resolving merge/rebase conflicts in refcache.json, see the resolve-refcache-conflicts skill.

References

Source-of-truth files — read on demand:

Overall Score

87/100

Grade

A

Excellent

Grades are signals, not a certification. Always review a skill yourself before use.

Safety

92

Quality

88

Clarity

85

Completeness

80

Summary

This skill guides an AI agent through a structured review workflow for pull requests in the open-telemetry/opentelemetry.io repository. It covers CI check verification, process rule validation (CLA, linked issues, co-ownership), content quality checks (front matter, terminology, links, markdown), and refcache handling. The skill is highly scoped to a single repository and defers blog-specific rules to a sibling skill.

Detected Capabilities

Read PR metadata (gh pr view)Read PR diff (gh pr diff)Check CI status (gh pr checks)Grep/search file contentClassify file pathsAnalyze Markdown and YAML front matterReference external documentation

Trigger Keywords

Phrases that agents use to match this skill to user intent.

review pull requestdebug ci failurepr checklistverify cla approvalcheck refcacheopentelemetry.io pr

Referenced Domains

External domains referenced in skill content, detected by static analysis.

opentelemetry.io

Use Cases

  • Reviewing a pull request in the open-telemetry/opentelemetry.io repository
  • Debugging a CI check failure on an opentelemetry.io PR
  • Preparing your own PR for submission to the repository
  • Verifying CLA coverage, linked issues, and co-ownership approvals
  • Validating front matter, terminology consistency, and internal link format
  • Checking refcache.json updates and external link handling

Quality Notes

  • Excellent scope definition: skill is narrowly scoped to a single repository (open-telemetry/opentelemetry.io) with clear boundaries and deferral to sibling skills for blog and issue drafting
  • Comprehensive process documentation: covers CLA, linked issues, co-ownership rules, locale span constraints, and branch freshness with inline citations to authoritative sources
  • Strong content review guidance: detailed checklist for front matter, terminology (e.g., 'OpenTelemetry' one word, proper casing of components), link format (internal refs vs. shortcut links), and markdown extensions
  • Well-structured workflow: five-step process (Setup → Walk CI → Process rules → Content review → Final pass) is logical and mirrors real PR review steps
  • Clear refcache handling guidance: explains when to use `?link-check=no` and defers complex merges to a dedicated skill (resolve-refcache-conflicts)
  • Appropriate use of allowed-tools: restricted to Bash, Read, Grep, Glob — no destructive or privilege-escalating commands
  • Excellent reference architecture: all authoritative sources (pr-checks.md, sig-practices.md, localization.md) are cited with deep links; agent knows to trust source docs when skill drifts
  • Practical argument parsing: handles multiple input formats (URL, #N, bare number) with clear error handling
  • Minor: Missing edge case documentation for empty PR diffs or repos with no open PRs, but this is minor given the skill's narrow scope
Model: claude-haiku-4-5-20251001Analyzed: Oct 29, 2026

Reviews

Add this skill to your library to leave a review.

No reviews yet

Be the first to share your experience.

Use fairoosansar-coder/review-pull-request in your dev environment

Command Palette

Search for a command to run...