CtrlK
BlogDocsLog inGet started
Tessl Logo

vibe-review

Review a proposed change against requirements, regressions, verification evidence, and relevant security risks.

60

Quality

70%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./templates/.agents/skills/vibe-review/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

86%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.

The body is an exemplary lean, instruction-only skill: it gives concrete inputs, defect categories, output format, honesty constraints about verification evidence, and clear scope limits with zero padding. The only minor improvements would be an example of a well-formed finding and slightly more explicit sequencing, but as a short single-purpose skill it is well above the bar.

DimensionReasoningScore

Conciseness

The ~9-line body is lean and assumes Claude's competence: no explanation of what a diff is, no library overviews, no padding. Every sentence gives a directive ('Read the actual diff...', 'Trace changed behavior into its callers and tests...', 'Report actionable findings with file/line, trigger, impact...'), so every token earns its place.

5 / 5

Actionability

For an instruction-only skill, the guidance is concrete: it names the exact inputs to read (diff, applicable instructions, acceptance criteria, check results), specific defect categories (missing error handling, incorrect assumptions, authorization or data exposure issues), and the output format (file/line, trigger, impact, suggested correction). Not 5 because there are minor gaps — no example of a well-formed finding, and 'applicable instructions' does not say where those instructions come from.

4 / 5

Workflow Clarity

The sequence is clear and ordered: gather inputs, trace changed behavior into callers and tests, hunt specific defect classes, then report with a defined format — plus explicit handling of the no-findings case ('If there are no findings, say so and disclose remaining verification gaps'). Checkpoints like 'Do not claim independent checks unless you ran them' and 'Distinguish demonstrated defects from missing evidence' act as verification gates. Not 5 because there is no explicit validate-a-finding-before-reporting loop or numbered step structure.

4 / 5

Progressive Disclosure

This is a single-file skill under 50 lines with no references/, scripts/, or assets/ directories — no external content is needed. Per the rubric's simple-skill exception, well-organized concise content alone merits a 5, and the two-paragraph structure (review process, then reporting rules and scope constraints) is clean and easy to navigate.

5 / 5

Total

18

/

20

Passed

Description

55%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.

The description states a clear and specific 'what' — reviewing a proposed change against requirements, regressions, verification evidence, and security risks — with no padding. Its main weaknesses are the complete absence of a 'when to use' trigger clause and the lack of the domain's most natural trigger terms (pull request, PR, diff), which also raises conflict risk with other review skills.

Suggestions

Add an explicit trigger clause, e.g. 'Use when reviewing a pull request, proposed change, or diff before merging' — this would lift completeness from 3 to 4-5.

Incorporate natural user vocabulary such as 'pull request', 'PR', 'diff', and 'code review' to improve trigger term quality and routing.

Sharpen distinctiveness by naming what sets this review apart (e.g. checking requirements/acceptance criteria compliance and verification evidence) so it is less likely to collide with generic code-review skills.

DimensionReasoningScore

Specificity

The description enumerates four concrete review dimensions ("requirements, regressions, verification evidence, and relevant security risks") beyond just naming the domain. It stops short of anchor 5 because it uses a single action verb ('Review') rather than multiple distinct concrete actions with comprehensive coverage.

4 / 5

Completeness

The 'what' is clear (review a proposed change against four criteria), but there is no 'Use when...' clause or equivalent trigger guidance — 'proposed' only weakly implies pre-merge timing. Per the rubric guideline, a missing 'when' caps completeness at 3; it is not 4 because the 'when' is absent rather than merely under-specified.

3 / 5

Trigger Term Quality

Terms like "review", "regressions", and "security risks" are natural, but the domain's most common user phrasings — 'pull request', 'PR', 'diff', 'code review' — are missing. This matches anchor 3 (some relevant keywords but missing common variations) rather than 4, which requires good coverage with only a few natural terms absent.

3 / 5

Distinctiveness Conflict Risk

"Review a proposed change" overlaps significantly with generic code-review and PR-review skills, so a user asking to 'review this PR' could route to the wrong skill. The requirements-compliance and verification-evidence emphasis provides some distinctiveness, matching anchor 3 rather than 4 (mostly distinct) or 2 (very broad).

3 / 5

Total

13

/

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

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

Total

15

/

16

Passed

Repository
KhazP/vibe-coding-prompt-template
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.