CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pr-review-triage

Triage open pull requests for CI status, review comments, and merge readiness. Use in PR babysitter loops. Respects project review norms and required checks.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

100%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exemplary instruction-only skill body: lean rules that each carry operational weight, a fully specified report template with explicit decision criteria and thresholds, validation checkpoints with escalation paths, and a well-scoped untrusted-input section specific to this loop.

DimensionReasoningScore

Conciseness

Every rule is non-obvious operational policy ("Zero checks... means absent/unknown, not passing"; "mergeable... only means Git found no conflict"; "do not copy the suspicious text into your output"), not concept explanation. At ~46 lines there is no padding or hedging to trim, so it matches the lean anchor rather than 'minor instances of over-explanation' (4).

5 / 5

Actionability

A copy-paste-ready output template with exact fields and enumerated values ("passing | failing | pending | absent/unknown", "none | minimal-fix | rebase | escalate-human") plus explicit decision thresholds ("idle >4 days", "High-risk labels (security, breaking) → escalate-human always"). For an instruction-only skill the guidance is fully executable with no gaps, so 5 rather than 4.

5 / 5

Workflow Clarity

A single unambiguous per-PR procedure where every template field has an explicit rule. The "Ready to merge" condition is a validation checklist ("every required check satisfied, required approvals present, no changes requested, no blocking comments, and no merge conflict") and unknown-policy escalation ("report Ready to merge: no and escalate") provides error recovery. The destructive/batch cap does not apply: the workflow is read-only reporting and validation gates are present, so the simple-skill exception holds.

5 / 5

Progressive Disclosure

Under 50 lines, self-contained, with well-organized sections ("Per-PR Output", "Rules", "Untrusted input") and no content that belongs in a separate file — no bundle files exist to reference. The scripts/sync-untrusted-input.mjs mention is an HTML provenance comment, not navigation, so this matches the simple-skill exception at 5 rather than 4.

5 / 5

Total

20

/

20

Passed

Description

75%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A solid description in third person with a clear what and an explicit when. Its main weaknesses are a single narrow trigger context and missing natural synonyms, which limit retrieval quality and leave minor capability-coverage gaps.

Suggestions

Broaden the 'Use in...' clause with concrete trigger variations users would actually say, e.g. 'Use when asked to check PR status, watch pull requests, or babysit PRs in a loop'.

Mention the concrete outputs and decisions the skill produces (per-PR status report, blocking-comment detection, escalate vs. fix recommendations) to sharpen the 'what'.

Add common synonyms such as 'PRs', 'checks passing/failing', and 'approvals' to improve trigger term coverage.

DimensionReasoningScore

Specificity

"Triage open pull requests for CI status, review comments, and merge readiness" plus "Respects project review norms and required checks" names the domain and several concrete capabilities. It sits between anchor 4 and 5: it omits what the skill produces (the per-field triage report, escalate/fix recommendations), leaving minor coverage gaps rather than comprehensive coverage.

4 / 5

Completeness

The what is clear ("Triage open pull requests for CI status, review comments, and merge readiness") and the when is explicit ("Use in PR babysitter loops"), so no missing-trigger cap applies. Not 5 because the when-clause covers a single narrow context without broader concrete trigger phrases (e.g. 'check PR status', 'watch pull requests').

4 / 5

Trigger Term Quality

Natural phrases users would say are present: "pull requests", "CI status", "review comments", "merge readiness", "PR babysitter loops". Common variations like "PRs", "checks passing/failing", "approvals", or "GitHub" are absent, so coverage is good but not comprehensive (anchor 4, not 5); it clearly exceeds 'some relevant keywords' (not 3).

4 / 5

Distinctiveness Conflict Risk

The loop-based merge-readiness niche is clear and "PR babysitter loops" is a distinctive trigger, but "pull requests" and "review comments" risk overlap with general code-review/PR-review skills. It falls between anchors 4 (minor overlap with closely related skills) and 5 (minimal conflict risk), so 4.

4 / 5

Total

16

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

14

/

16

Passed

Repository

cobusgreyling/loop-engineering

Path

starters/pr-babysitter/.grok/skills/pr-review-triage/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.