CtrlK
BlogDocsLog inGet started
Tessl Logo

pr-review

Use when authoring a code review of a pull request or merge request — "review this PR", "do a code review on PR

71

Quality

86%

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

77%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.

A well-architected orchestration skill with a clear, validated multi-step workflow and clean one-level-deep progressive disclosure. Its weakness is verbosity: rationale and failure anecdotes inflate the body without adding executable value.

Suggestions

Tighten the prose rationale into terse imperatives — e.g. replace 'A review is worth only as much as its independence, and the merge is where independence quietly dies: ...' with 'Triage in a fresh subagent so the merge verdict isn't biased by this session's own reasoning.'

Move the observed-failure anecdotes ('three sessions running', 'six consecutive amendments') into a short 'Why this gate exists' note in references/house-style.md rather than inline in the workflow steps.

Collapse the repeated emphasis on brevity-modes-never-shrink-deliverables (stated in step 5, step 7, and Boundaries) into a single Boundaries bullet the steps reference.

DimensionReasoningScore

Conciseness

Mostly efficient operational guidance, but padded with rationale and observed-failure anecdotes that could be tightened — e.g. 'A review is worth only as much as its independence, and the merge is where independence quietly dies...', 'The documented failure of this step (three sessions running)', 'The failure this prevents was observed across six consecutive amendments'.

3 / 5

Actionability

Concrete, executable guidance throughout: named operations (preflight, fetch-pr-context, create-draft-review), specific commands (gh auth status, glab auth status, git remote get-url origin), a concrete draft path review/pr-<N>-draft.md, and a specified subagent_type; minor gaps because the operational detail lives in the referenced files rather than inline.

4 / 5

Workflow Clarity

A clearly sequenced 7-step workflow with explicit validation checkpoints (the step-5 results gate, the pre-gate protocol, the preflight delivery-capability check, the never-destroy rule) and feedback loops (evidence-pass re-present, re-review loop, amend-in-place), satisfying the destructive/outward-facing validation requirement.

5 / 5

Progressive Disclosure

The body is a well-signaled overview pointing one level deep to real reference files ([references/analysis.md], [references/github.md], [references/gitlab.md], [references/local.md], [references/house-style.md]) — all of which exist in ./references/ — with content appropriately split and easy navigation.

5 / 5

Total

17

/

20

Passed

Description

95%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 strong, specific description with comprehensive trigger terms, clear what/when coverage, and low conflict risk. Its only weakness is second-person voice ('your', 'you'), which the rubric penalizes on specificity.

DimensionReasoningScore

Specificity

Names the domain and a comprehensive set of concrete actions ('reviews someone else's PR/MR', 'posts the result as a draft review', 'reviews your own working branch and writes the review to a file', 'orchestrating the code-review and pr-review-toolkit plugins', 'gates everything on your approval'), which is a 5-anchor fit; reduced by 1 per the rubric's second-person voice penalty ('your approval', 'your own working branch', 'a PR you authored').

4 / 5

Completeness

Explicitly answers both what (authors a code review, finds issues via plugins, drafts in house style, gates on approval) and when ('Use when authoring a code review of a pull request or merge request') with concrete trigger phrases.

5 / 5

Trigger Term Quality

Comprehensive natural trigger phrases including synonyms and variations: 'review this PR', 'do a code review on PR #N', 'review this MR', 'review my branch', 'leave review comments', plus local-mode triggers 'locally', 'for myself', 'just my branch', 'don't post'.

5 / 5

Distinctiveness Conflict Risk

Clear niche as 'the reviewer's side' that distinguishes it from sibling skills, with an explicit hand-off ('to respond to feedback on a PR you authored, use pr-comments-address') and distinct, specific triggers that minimize overlap.

5 / 5

Total

19

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
provectus/awos-recruitment
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.