agentleFS
Sign inStart free
Goodeye-LabsGoodeye-Labs/truesight-mcp-skills/skills/eval-audit/SKILL.md

eval-audit

Skill7 starsChanged 8 months agoMITView on GitHub

Audit an existing evaluation workflow and produce severity-ranked findings with concrete next actions. Use when inheriting an eval setup, diagnosing quality regressions, or checking LLM evaluation process maturity.

5 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use validate-evaluator and write-judge-prompt.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • tim-osterhusSkill

    validate-evaluator

    Guardrail for validating LLM judges against human labels using data splits, TPR/TNR, disagreement review, and bias correction.

    Also in 3 projects that use this

  • eggboySkill

    write-judge-prompt

    Design LLM-as-Judge evaluators for subjective criteria that code-based checks cannot handle. Use when a failure mode requires interpretation (tone, faithfulness, relevance, completeness). Do NOT use when the failure mode can be checked with code (regex, schema validation, execution tests). Do NOT use when you need to validate or calibrate the judge — use validate-evaluator instead.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    BEFORE the first scoping question, search for a structured question tool (e.g., AskUserQuestion or similar interactive widget) and load it.

    line 13
  2. Distinctive:

    Ask one question at a time using the structured question tool (loaded per the HARD-GATE above).

    line 16
  3. Distinctive:

    Use the structured question tool for every question.

    line 30
  4. Distinctive:

    Ask one follow-up only if ambiguity remains.

    line 31
  5. Distinctive:

    If evidence is missing, record that as a finding.

    line 42
  6. Distinctive:

    Keep scope within current Truesight MCP capabilities.

    line 83

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Eval Audit

Audit LLM evaluation practice and route gaps to the right skills.

Interactive Q&A protocol (mandatory)

<HARD-GATE> BEFORE the first scoping question, search for a structured question tool (e.g., `AskUserQuestion` or similar interactive widget) and load it. Use that tool for EVERY scoping question. Fall back to plain-text lettered options ONLY if no such tool exists in the environment. </HARD-GATE>

Ask one question at a time using the structured question tool (loaded per the HARD-GATE above).

Example question structure:

What should this audit prioritize first?
A) Live evaluation quality and coverage
B) Error analysis maturity
C) Review and promotion loop health
D) End-to-end process health

Rules:

  • One question per message.
  • Use the structured question tool for every question. Structure each with a short header, 2-4 options with labels and descriptions, and place the recommended option first. Do not add "(Recommended)" or similar annotations to option labels.
  • Ask one follow-up only if ambiguity remains.

Inputs and evidence

Collect available evidence from Truesight first:

  • datasets and dataset rows
  • live evaluations
  • evaluation runs/results
  • review queue items
  • existing evaluation criteria and deployment patterns

If evidence is missing, record that as a finding.

Diagnostic areas

  1. Evaluation coverage and quality dimensions
  2. Error analysis practice and category quality
  3. Review and promotion workflow discipline
  4. Template usage versus custom needs
  5. Operational hygiene (verification, reruns, iteration cadence)

Report format (mandatory)

For each finding, include:

### <Finding title>
Status: Problem exists | OK | Cannot determine
Evidence: <specific evidence from Truesight context>
Severity: critical | high | medium | low
Recommended skill: <one of current skill set>
Next command: <concrete instruction to run next>

Order findings by severity and impact.

Severity rubric

  • critical: likely causes incorrect go/no-go decisions or severe user harm
  • high: frequent quality failures or missing control loops
  • medium: meaningful process weakness with moderate impact
  • low: optimization opportunity, documentation, or ergonomics issue

Handoff map

  • Missing or weak failure taxonomy -> error-analysis
  • Missing live evaluation coverage -> create-evaluation or bootstrap-template-evaluation
  • Review backlog or low judgment throughput -> review-and-promote-traces
  • Unclear starting path -> truesight-workflows

Guardrails

  • Keep scope within current Truesight MCP capabilities.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free