Use when the user wants to build an evaluation system for an LLM/agent application but doesn't know where to start — they have traces, prompts, RAG pipelines, or nothing at all. Also use when the user mentions evaluation, eval, benchmarking, testing LLM quality, measuring agent performance, assessing RAG accuracy, or wants to compare prompts/models. This skill is the entry router: it asks diagnostic questions then recommends which sub-skill (local workflow) to use next.
SkillScore86Maintained
Automatic
◆AI Engineer pack
Last verified 8/15/2026
What it changes
Passed 19 of 22 test cases with the skill, and 3 of 22 without it.
Without skill14%
With skill86%
Checks it passed that the model alone missed
The response recommends 07-redteam as the workflow.
The response recommends 03-align-human as the workflow.
The response recommends 01-eval-design as the workflow.
and 24 more in the full report
17 fixed2 passed either way2 still failing1 broke
Pass-rate lift
+73 pp
Tokens
+94%
Agent turns
0%
Response time
-32%
Model gemini-3.6-flashMeasured 8/15/2026v119 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 17 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 17
Prompt
Please triage this situation: I have design spec docs, zero labels, low internal stakes, and good domain knowledge. Format your answer exactly using the standard meta-eval output block.
What was checked, without and with the skill
Checked
Without skill
With skill
The response formats the Diagnosis summary line using pipe-separated data, labels, stakes, and domain keys.
Missed
Passed
The response recommends 01-eval-design as the entry workflow.
Missed
Passed
The response includes 'Why:' and 'What this workflow will do:' sections in the output.
Missed
Passed
case-22 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md182 linessingle file
Meta Eval
Entry router for the eval skill collection. You diagnose what the user has and route
them to the right sub-skill. You don't do evaluation yourself — you're the triage desk.
Each sub-skill is self-contained: it carries inline the data shapes, statistics, and data
principles it needs, so it can be installed and used on its own.
Checklist
You MUST create a task for each item and complete them in order:
Match triage table — map user scenario to sub-skill
Recommend sub-skill — tell the user which workflow to use and why
Record routing decision — write a brief summary of what was diagnosed and recommended
Diagnostic Questions
Ask these 4 questions (all at once — don't drip-feed):
To route you to the right evaluation skill, I need to understand your situation:
1. What data do you have?
a) Agent traces / production logs
b) Product spec / design docs
c) Nothing yet — starting from scratch
2. Do you have human labels?
a) Yes, ≥50 labeled examples
b) Some, but fewer than 50
c) None
3. What are the stakes?