eval-measure

eval-measure is a skill for Claude Code, Codex from Galileo-Agent-Labs/eval-engineer. It costs 36 tokens per session (423 once invoked), scanned A, original, MIT.

A measurement guide for checking whether an AI application produces the expected results safely and accurately. It defines what a correct answer must contain, what it must avoid, and how to assess it.

In plain words
What is it for?
Use it to create evaluation rules, expected-output contracts, quality measurements, and acceptance gates before changing an AI application.
Why use it?
It prevents teams from optimizing an AI system without knowing whether it is actually improving. It also exposes missing checks for sources, permissions, safety, and answer quality.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to create evaluation rules, expected-output contracts, quality measurements, and acceptance gates before changing an AI application.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/galileo-agent-labs/eval-engineer/eval-measure
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Galileo-Agent-Labs/eval-engineer --skill eval-measure
Clone the repo
git clone --depth 1 https://github.com/Galileo-Agent-Labs/eval-engineer

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for eval-measure

README.md
[![agentmods](https://agentmods.dev/badge/skills/galileo-agent-labs/eval-engineer/eval-measure.svg)](https://agentmods.dev/skills/galileo-agent-labs/eval-engineer/eval-measure)
Your own site
<a href="https://agentmods.dev/skills/galileo-agent-labs/eval-engineer/eval-measure"><img src="https://agentmods.dev/badge/skills/galileo-agent-labs/eval-engineer/eval-measure.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 423 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00036 $0.00423
Opus 5 $0.00018 $0.00211
Sonnet 5 $0.00007 $0.00085
Haiku 4.5 $0.00004 $0.00042

Measured 8d ago against content hash 058efcb6ed5f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

eval-measure scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/eval-measure/SKILL.md · 52 lines

What it actually says

Eval Measure

Use this skill before optimization or broad fixture work. Its job is to make the measurement contract explicit.

Required Reference

Use skills/eval-engineer/references/metric-profile-checklist.md and skills/eval-engineer/assets/metric-profile-template.md.

Do

  • Define risk profile and quality dimensions.
  • Write the full expected-output contract: expected decision, required and forbidden citations, tools, answer constraints, abstention, permissions, and safety requirements.
  • Include retrieved-source gates when source authority matters: required_retrieved_sources, forbidden_retrieved_sources, and whether final citations alone are sufficient for the case risk.
  • Prefer independent observations over agent self-reports for safety gates. Check answer text, citations, retrieved source IDs, tool calls, and Galileo scorers before accepting flags emitted by the app under test.
  • Choose Galileo metrics by failure contract, not by one global list.
  • Identify metric gaps before accepting a cost or quality change.

Gotchas

  • Agent self-reported flags are instrumentation breadcrumbs, not acceptance evidence.
  • Final citations alone are not enough when retrieved-source authority is part of the risk.
  • Do not infer latency, wall time, tokens, or cost as quality metrics.

Validation Loop

Before finalizing a metric profile, check it against skills/eval-engineer/references/metric-profile-checklist.md and confirm the profile names quality gates, cost/performance metrics, segment gates, metric direction, and known gaps.

Output

Findings first. Start with the highest-risk metric gaps and the keep/reject/ inconclusive measurement decision, then produce a metric profile or explain what evidence is missing. Do not improve the app until the expected-output contract and acceptance gates are clear.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 52 lines · 36 tokens per session scan A 058efcb6ed5f

Subscribe to this mod's changes

eval-measure is a skill published in the GitHub repository Galileo-Agent-Labs/eval-engineer (41 stars, last pushed 27d ago), licensed MIT. It adds 36 tokens to every session and 423 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

rag-evaluation

Comprehensive RAG evaluation with retrieval metrics, generation quality, and end-to-end testing. Use this skill when measuring and improving RAG system performance. Activate when: RAG evaluation, RAGAS, retrieval metrics, generation quality, RAG testing, MRR, recall, faithfulness.

latestaiagents/agent-skills · 63 tokens

llm-fallback-chains

Use this skill when implementing fallback strategies for LLM applications. Activate when the user needs graceful degradation for AI services, multi-provider failover, handling LLM outages, or building resilient AI systems.

latestaiagents/agent-skills · 52 tokens

agentic-rag

Build autonomous RAG agents that reason, plan, and use tools for complex retrieval tasks. Use this skill when simple retrieve-and-generate isn't enough. Activate when: agentic RAG, RAG agent, multi-step retrieval, tool-using RAG, autonomous retrieval, query decomposition.

latestaiagents/agent-skills · 63 tokens

corrective-rag

Implement Corrective RAG (CRAG) with retrieval validation, fallback strategies, and self-correction. Use this skill when RAG outputs need quality guarantees and automatic error correction. Activate when: CRAG, corrective RAG, retrieval validation, fallback search, self-correcting RAG, grounded generation.

latestaiagents/agent-skills · 67 tokens

graphrag-patterns

Implement GraphRAG patterns combining knowledge graphs with retrieval for complex reasoning. Use this skill when building RAG over interconnected data or needing relationship-aware retrieval. Activate when: GraphRAG, knowledge graph, graph retrieval, entity relationships, Neo4j RAG, graph database, connected data.

latestaiagents/agent-skills · 65 tokens

production-rag-checklist

Comprehensive checklist for deploying RAG systems to production with reliability and scale. Use this skill when preparing RAG for production deployment. Activate when: production RAG, RAG deployment, RAG checklist, RAG scaling, RAG monitoring, production-ready RAG.

latestaiagents/agent-skills · 60 tokens