CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gsd-eval-review

Audit an executed AI phase's evaluation coverage and produce an EVAL-REVIEW.md remediation plan.

Quality

71%

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 ./pi/package-skills/gsd-eval-review/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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 a lean, platform-aware wrapper: it gives concrete dispatch and include-expansion instructions with explicit fail-closed error handling, and defers the detailed workflow to clearly signaled one-level-deep references. It scores consistently at 4 — solid executable guidance with minor gaps around verification of the deferred workflow content.

DimensionReasoningScore

Conciseness

The body is dense, operational, and free of concept explanations Claude already knows; the only near-padding is the '<arguments>' data-vs-instructions note, which is justified as injection-safe handling, leaving minor tightening opportunities — the anchor 4 'efficient; minor instances of over-explanation' rather than the fully lean anchor 5.

4 / 5

Actionability

Gives concrete, executable directives ('call gsd_dispatch with role ... and task', 'read every file listed under <execution_context>', 'STOP that step and surface the error', 'Never execute the role's work inline') with concrete error-handling branches; not 5 because the core workflow steps themselves are delegated to the referenced engine files rather than exemplified inline.

4 / 5

Workflow Clarity

Sequence is explicit ('Before doing anything else, read every file... then follow the workflow they define') with fail-closed error checkpoints (unresolved role or failed child -> stop, surface, offer skip if optional); not 5 because the multi-step process is largely defined out-of-body and 'Execute end-to-end. Preserve all workflow gates.' is a bare directive rather than a checklist.

4 / 5

Progressive Disclosure

A short, well-sectioned wrapper that clearly signals two one-level-deep references under <execution_context> and explicitly handles the platform's include limitation; scored 4 rather than 5 because the referenced files are placeholder engine-root paths not present as verifiable bundle files, a minor organization gap.

4 / 5

Total

16

/

20

Passed

Description

53%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 concrete, moderately specific capability with a clear output artifact in third person, but it entirely lacks 'when to use' trigger guidance and natural trigger phrases. It is serviceable but below the bar set by the good examples, which all pair the 'what' with an explicit 'Use when...' clause.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user asks to review or audit the evaluation coverage of a completed AI phase, or requests an EVAL-REVIEW.md remediation plan.'

Include natural trigger terms and synonyms users would actually say, such as 'eval review', 'evaluation audit', 'test coverage check', or 'AI-SPEC.md evaluation strategy'.

Expand the 'what' slightly to enumerate what the audit produces (score, verdict, gaps, remediation plan) to lift specificity toward comprehensive coverage.

DimensionReasoningScore

Specificity

Names the domain ('evaluation coverage' of 'an executed AI phase') and two concrete actions ('Audit... evaluation coverage', 'produce an EVAL-REVIEW.md remediation plan') with a specific output artifact, but coverage is not comprehensive — matching the 1-2 concrete actions anchor rather than the 'several specific actions' anchor.

3 / 5

Completeness

The 'what' is clear (audit evaluation coverage, produce an EVAL-REVIEW.md remediation plan), but there is no 'Use when...' clause or any explicit trigger guidance for when to invoke the skill, which caps completeness at 3 per the judging guidelines; not 2 because the 'what' is concrete rather than vague.

3 / 5

Trigger Term Quality

Contains relevant keywords ('evaluation coverage', 'remediation plan', 'EVAL-REVIEW.md', 'AI phase') but misses common natural variations a user would say, such as 'eval review', 'audit the evals', or 'evaluation strategy', matching the 'some relevant keywords but missing common variations' anchor.

3 / 5

Distinctiveness Conflict Risk

The GSD-specific framing ('executed AI phase') and named artifact 'EVAL-REVIEW.md' give it a mostly distinct niche with only minor overlap risk against generic review/audit skills; not 5 because no explicit distinct trigger phrases are provided to minimize conflict.

4 / 5

Total

13

/

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

open-gsd/gsd-core

Path

pi/package-skills/gsd-eval-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.