When reviewing, optimizing, or inheriting an LLM agent, systematically identify what's actually load-bearing vs. cruft. Instruments usage, ablates components, and strips the agent to minimum viable form. Answers: "What can we remove without degrading quality?" The goal is the leanest agent that still works.
SkillScore77Maintained
Automatic
◆AI Engineer pack
Last verified 8/18/2026
What it changes
Passed 13 of 17 test cases with the skill, and 9 of 17 without it.
Without skill50%
With skill77%
Checks it passed that the model alone missed
Recommends consolidating the focus areas down into the target range of 5 to 15 specific items.
Evaluates the 250-line size as excellent and concise, requiring no progressive disclosure split.
Flags the count of 2 focus areas as a critical defect because at least 3 focus areas are required.
and 4 more in the full report
7 fixed10 passed either way4 still failing1 broke
Pass-rate lift
+27 pp
Tokens
+93%
Agent turns
0%
Response time
-34%
Model gemini-3.6-flashMeasured 8/18/2026v217 of 22 cases passed with skill · 17 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 7
Prompt
An agent definition is 250 lines long and contains no reference directory. Assess whether this agent needs to be split into multiple reference files.
What was checked, without and with the skill
Checked
Without skill
With skill
Evaluates the 250-line size as excellent and concise, requiring no progressive disclosure split.
Missed
Passed
case-15 · gemini-3.6-flash · 8/18/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md379 linessingle file
Reference Files
Advanced agent validation guidance:
model-selection.md - Model choice decision matrix, use cases, and appropriateness criteria
tool-restrictions.md - Tool permission patterns, security implications, and restriction fit
focus-area-quality.md - Focus area specificity assessment, quality scoring, and criteria
approach-methodology.md - Approach completeness, required components, and methodology patterns
resource-organization.md - Resource directory validation and progressive disclosure
examples.md - Good vs poor agent comparisons and full audit reports
report-format.md - Standardized audit report template and structure
common-issues.md - Frequent problems, fixes, and troubleshooting patterns
Agent Auditor
Validates agent configurations for model selection, tool restrictions, focus areas, and approach methodology.
Quick Start
Basic audit workflow:
Read agent file
Check model selection appropriateness
Validate tool restrictions
Assess focus area quality
Review approach methodology
Generate audit report
Example usage:
User: "Audit my evaluator skill"
→ Reads skills/evaluator/SKILL.md
→ Validates model (Sonnet), tools, focus areas, approach
→ Generates report with findings and recommendations