CtrlK
BlogDocsLog inGet started
Tessl Logo

uinaf/skill-audit

Audit existing skills with Tessl scoring, metadata and trigger-coverage checks, repo conventions, and skill-authoring best practices. Use when creating or revising a skill, triaging weak self-activation, or comparing a skill against source-repo guidance such as `AGENTS.md`, `CLAUDE.md`, or repo rules, plus external skill guidance. Do not use to verify general application code or to rewrite unrelated docs.

97

1.05x
Quality

98%

Does it follow best practices?

Impact

97%

1.05x

Average score across 3 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

Overview
Quality
Evals
Security
Files

Evaluation results

96%

2%

Skill Portfolio Review: Activation and Routing Problems

Criteria
Without context
With context

Tessl run per skill

100%

100%

Score per skill reported

100%

100%

Vague names flagged

50%

50%

Missing boundaries flagged

100%

100%

Third-person descriptions proposed

100%

100%

What and when in descriptions

100%

100%

Overlap boundary in descriptions

100%

100%

Misplaced code review content flagged

100%

100%

Correct handoff skill named

77%

100%

Evidence over taste

100%

100%

Optimizer not invoked

100%

100%

Scope respected

100%

100%

100%

5%

Skill Health Check: Release Coordinator

Criteria
Without context
With context

Rubric moved to references

100%

100%

Examples moved to references

100%

100%

Duplicate checklist removed

100%

100%

Frontmatter cleaned

44%

100%

References linked from SKILL.md

100%

100%

Tessl run before edits

100%

100%

Tessl rerun after edits

100%

100%

Before/after scores compared

100%

100%

Change reasons documented

100%

100%

SKILL.md remains workflow-focused

100%

100%

Optimizer not used

100%

100%

Scope respected

100%

100%

95%

6%

Skill Audit: Onboarding Assistant Skill

Criteria
Without context
With context

Tessl run first

100%

100%

Correct single-skill command

100%

100%

Score reported

100%

100%

Tessl command reported

100%

100%

Strongest parts identified

100%

100%

Findings with file references

100%

100%

Blockers separated from polish

66%

100%

Smallest change set

100%

100%

Rerun status reported

100%

100%

Repo conventions loaded

71%

71%

Optimizer not invoked

100%

100%

Adjacent issues flagged separately

33%

66%

Evaluated
Agent
Claude
Model
Claude Sonnet 4.6

Table of Contents