CtrlK
BlogDocsLog inGet started
Tessl Logo

code-audit

Codebase-specific smell catalog with calibrated grep recipes: duck-typing at boundaries, yargs argv re-assertions, value-discarding promise tails, console.* in library code, exhaustive `never` over-broad unions, single-method `Pick` deps, `*Like` boundary shape copies. Use for a periodic audit pass, scoping a cleanup PR, after a major refactor, or reviewing a primitive's consumers. Pairs with refactoring (fix mechanics) and post-implementation-review (second-read hub).

80

Quality

100%

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

SKILL.md
Quality
Evals
Security

Quality

Content

100%

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 tight, actionable audit catalog: each smell has an executable detection recipe, triage thresholds, and false-positive guards, capped by a clear sequenced workflow. It assumes Claude's intelligence and avoids both verbosity and concept re-explanation.

DimensionReasoningScore

Conciseness

The body is dense signal throughout — pattern, executable grep, 'why it matters', triage, and false positives per category — and assumes Claude's competence without explaining TypeScript, yargs, or grep. Not a level-2 case: there is no padded conceptual explanation, only earned, codebase-specific detail.

3 / 3

Actionability

Every category ships a copy-paste-ready grep/rg recipe plus concrete triage thresholds ('Five flags or three commands sharing the same accessor pattern', '5+ is a smell'), fully executable rather than pseudocode.

3 / 3

Workflow Clarity

A dedicated 'Workflow' section sequences the pass (run greps -> triage each hit as justified/refactor/false-positive -> group findings -> document in one PR), with an explicit classification checkpoint; the read-only audit nature means the destructive-operation validation cap does not apply.

3 / 3

Progressive Disclosure

No bundle files exist, and all content is inline with well-organized numbered sections and clearly signaled cross-skill pointers (post-implementation-review, refactoring, define-errors) rather than nested file references; per the rubric's simple-skill note, well-organized sections without external references score 3.

3 / 3

Total

12

/

12

Passed

Description

100%

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 is specific, complete, and well-differentiated: it lists concrete actions, gives explicit use-when triggers, and clarifies its boundaries against related skills. Voice is appropriately third person and there is no vague fluff.

DimensionReasoningScore

Specificity

The description enumerates seven concrete, named smell-detection actions ("duck-typing at boundaries", "yargs argv re-assertions", "value-discarding promise tails", "console.* in library code", "exhaustive `never` over-broad unions", "single-method `Pick` deps", "`*Like` boundary shape copies"), matching the 'lists multiple specific concrete actions' anchor.

3 / 3

Completeness

It clearly answers both what ("Codebase-specific smell catalog with calibrated grep recipes") and when ("Use for a periodic audit pass, scoping a cleanup PR, after a major refactor, or reviewing a primitive's consumers"), with an explicit trigger clause.

3 / 3

Trigger Term Quality

Natural trigger phrasing a user would say is well covered ("periodic audit pass", "scoping a cleanup PR", "after a major refactor", "reviewing a primitive's consumers"); while some catalog terms are technical, the trigger clause uses the domain's real vocabulary.

3 / 3

Distinctiveness Conflict Risk

It carves a clear niche (codebase-specific smell auditing) and explicitly distinguishes it from adjacent skills via 'Pairs with refactoring (fix mechanics) and post-implementation-review (second-read hub)', making wrong-skill triggering unlikely.

3 / 3

Total

12

/

12

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
EpicenterHQ/epicenter
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.