CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

code-review

Review a diff for proven defects. Trigger: review this PR/branch/diff, code review, pre-merge check.

Quality

85%

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

92%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 skill body: lean, opinionated, and sequenced with explicit done-when checkpoints and a verify-then-downgrade feedback loop, with detail correctly pushed to two real, one-level-deep reference files. The only weakness is minor executability gaps in the action guidance.

DimensionReasoningScore

Conciseness

The ~30-line body contains no explanations of concepts Claude already knows and no padding; every line is instruction or a standard ("A finding with a failure scenario is a defect; without one it is a preference — drop it or mark it a nit"). This matches the lean, every-token-earns-its-place anchor.

5 / 5

Actionability

Concrete, copy-paste commands are given for each platform ("git diff <base>...HEAD", "gh pr diff <id>", "ado pr diff <id> --hunks") and the output format is fully specified ("One finding per line: **what — where (`file:line`) — failure scenario — fix**"). Minor gaps keep it below anchor 5: how to determine the diff base is left implicit, the verify step's checks are conditional ("Run the relevant test or a read-only check when you can"), and no rendered example finding appears inline.

4 / 5

Workflow Clarity

Five clearly sequenced steps each end in an explicit "*Done when*" checkpoint, and step 4 is a feedback loop ("try to prove it wrong... Keep what survives; set confidence from what you actually observed"). This matches the anchor requiring explicit validation steps and feedback loops; the destructive/batch cap does not apply since the skill is read-only by design.

5 / 5

Progressive Disclosure

Both referenced files (references/lenses.md, references/findings.md) exist, are exactly one level deep, and are clearly signaled with their purpose ("see references/findings.md for the bar"). The bulky checklist content (lenses) is correctly split out of the overview, and the body itself is well organized — matching the clear-overview anchor.

5 / 5

Total

19

/

20

Passed

Description

73%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 tight, honest description with an explicit trigger clause and natural trigger phrases — better than most. Its main limitation is a thin "what": one action with no indication of what the review produces, which caps specificity and completeness.

Suggestions

Name 2-3 concrete actions in the description (e.g., "trace callers of changed code, verify each candidate defect, rank findings by severity") to raise specificity beyond a single action.

State the output briefly in the "what" (e.g., "...producing ranked one-line findings with failure scenarios and fixes") so the skill's result is explicit rather than inferred.

Add natural trigger variations users commonly say, such as "review my changes", "review this commit", or "pre-merge review", to broaden keyword coverage.

DimensionReasoningScore

Specificity

"Review a diff for proven defects" names the domain and a single concrete action but lists no further actions or outputs. This matches the anchor for 1-2 concrete actions without comprehensive coverage; it is not score 4 because it does not enumerate several specific actions.

3 / 5

Completeness

Both parts are present: the "what" ("Review a diff for proven defects") and an explicit "Trigger:" clause with concrete trigger phrases, satisfying the Use-when requirement. It sits between anchors 4 and 5 — the "when" is anchor-5 quality, but the "what" is minimal (one action, no mention of the output the skill produces), so it does not clearly match anchor 5.

4 / 5

Trigger Term Quality

"review this PR/branch/diff, code review, pre-merge check" are natural phrases a user would actually say. It falls short of anchor 5 only because common variations such as "review my changes", "review this commit", or "review the diff" are absent.

4 / 5

Distinctiveness Conflict Risk

The description claims a clear niche (diff/PR review for proven defects) with distinct, unambiguous triggers ("review this PR/branch/diff, code review, pre-merge check"), matching the anchor-5 example's structure. Minor overlap with a security-review skill exists but the triggers do not collide.

5 / 5

Total

16

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

YoizenSA/dev-ai-workflow

Path

ywai/skills/code-review/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.