CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gsd-audit-fix

Autonomous audit-to-fix pipeline — find issues, classify, fix, test, commit

Quality

72%

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-audit-fix/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

76%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, disciplined dispatcher document: it gives concrete fail-closed rules for Pi dispatch, includes, arguments, and flags without a wasted token. Its weaknesses are the delegation itself — the multi-step fix/test/commit workflow and its validation checkpoints exist only in an external file the body never summarizes, and the single reference uses a templated, unresolvable path.

Suggestions

Add a brief explicit validation loop for the batch-fix cycle in the body (fix → run tests → only commit on green tests, abort and revert on failure) so workflow clarity is not solely delegated to the external file.

Summarize what the referenced workflow (gsd-core/workflows/audit-fix.md) covers — its main phases and dispatch roles — so the body works as a true overview rather than a bare pointer.

Note how __GSD_PI_ENGINE_ROOT__ is resolved at install time, or replace it with a concrete relative path, so the single external reference is verifiable and navigable.

DimensionReasoningScore

Conciseness

The ~30-line body contains only environment-specific operational rules (fail-closed dispatch semantics, include expansion, argument-data handling, flag defaults) and explains nothing Claude already knows. Every token earns its place, matching anchor 5; anchor 4 would require trimmable over-explanation, which is absent.

5 / 5

Actionability

Concrete, executable guidance is present: call gsd_dispatch with role and task, read every execution_context file first, flags with explicit defaults (--max N default 5, --severity, --dry-run, --source), and explicit stop-on-error behavior. It falls short of 5 because the actual fix/test/commit mechanics are entirely deferred to the external workflow file, leaving a gap in the body itself.

4 / 5

Workflow Clarity

This is a batch/destructive-adjacent skill (autonomous fixes with commits), and the body's only validation checkpoint is the fail-closed dispatch error handling — the fix/test/commit cycle's verification steps are not written out, only implied by "test verification" in the objective and "Execute end-to-end". Per the guideline capping batch-operation skills without explicit validation at 3, this matches anchor 3 rather than 4.

3 / 5

Progressive Disclosure

The short body delegates its one external dependency clearly and one level deep under <execution_context> with an explicit instruction to read it first, and there are no bundle files to organize. It misses 5 because the reference path is an unresolved template variable (__GSD_PI_ENGINE_ROOT__) and the body offers no overview of what the referenced workflow contains, a minor navigation gap matching anchor 4.

4 / 5

Total

16

/

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 communicates a concrete five-action pipeline in few words, but entirely omits when to invoke it and what domain it operates on. It reads as a capability label rather than a trigger-ready skill description.

Suggestions

Add an explicit "Use when..." clause with concrete trigger phrases (e.g., "Use when the user asks to audit a codebase and automatically fix findings, run audit results to resolution, or batch-fix issues") to lift the completeness cap.

State the domain of "issues" (e.g., code/repository audit findings) so the actions are specific and distinguishable from generic lint or code-review skills.

Include natural synonyms users would say — bugs, lint errors, code review findings, auto-fix — to broaden trigger-term coverage.

DimensionReasoningScore

Specificity

"find issues, classify, fix, test, commit" names five concrete pipeline actions, matching the 'several specific actions; minor gaps' anchor. It falls short of 5 because the target of the actions ("issues") is generic — the domain (code, repo, infra) is never stated.

4 / 5

Completeness

The "what" is clear (find, classify, fix, test, commit), but there is no "Use when..." clause or any equivalent explicit trigger guidance, which per the judging guidelines caps completeness at 3.

3 / 5

Trigger Term Quality

Terms like "audit", "fix", and "commit" are phrases users naturally say, but common variations and synonyms (bugs, lint, code review, repair, CI failures) are missing, matching anchor 3 rather than the near-complete coverage of anchor 4.

3 / 5

Distinctiveness Conflict Risk

"Autonomous audit-to-fix pipeline" carves out a niche, but without distinguishing triggers it could still overlap with code-review, lint-fix, and general bug-fixing skills, matching anchor 3; it is not generic enough for 2 nor distinct enough for 4.

3 / 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-audit-fix/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.