agent-evaluation

agent-evaluation is a skill for Claude Code, Codex from MLOps-Courses/agentops-open-course. It costs 71 tokens per session (728 once invoked), scanned A, original, MIT.

A testing approach for coding agents that checks their actions and evidence across fixed cases, rather than judging only one final answer. It covers tool-use checks, groundedness, token changes, and prompt comparisons.

In plain words
What is it for?
Evaluate prompt or model changes, compare two agent versions, check that claims come from retrieved evidence, and turn real failures into regression tests.
Why use it?
An agent can appear correct while using the wrong tools, making unsafe changes, using unsupported claims, or becoming more expensive. Repeatable checks make those problems visible.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/mlops-courses/agentops-open-course/agent-evaluation
Any agent
npx skills add MLOps-Courses/agentops-open-course --skill agent-evaluation
Clone the repo
git clone --depth 1 https://github.com/MLOps-Courses/agentops-open-course

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 agent-evaluation

README.md
[![agentmods](https://agentmods.dev/badge/skills/mlops-courses/agentops-open-course/agent-evaluation.svg)](https://agentmods.dev/skills/mlops-courses/agentops-open-course/agent-evaluation)
Your own site
<a href="https://agentmods.dev/skills/mlops-courses/agentops-open-course/agent-evaluation"><img src="https://agentmods.dev/badge/skills/mlops-courses/agentops-open-course/agent-evaluation.svg" alt="Measured on agentmods" height="20"></a>
Per session 71 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 728 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00071 $0.00728
Opus 5 $0.00036 $0.00364
Sonnet 5 $0.00014 $0.00146
Haiku 4.5 $0.00007 $0.00073

Measured 3d ago against content hash 1539d25490ca, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-evaluation 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 3d 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/agent-evaluation/SKILL.md · 37 lines

How it starts

The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Evaluation

Score an agent's behavior over fixed cases, not one exact string. Let model-free structure checks gate merges. Treat every model-backed scorer as evidence with an explicit floor, and never make an LLM judge the sole release criterion.

When to use

  • A prompt/model change can pass a smoke test yet call the wrong tools, cost more, or hallucinate.
  • You want deterministic evalset validation in pull requests and behavioral evidence before release.
  • You need to choose between two prompt versions with numbers, not opinion.

Steps

  1. Score the trajectory, not the wording. Assert which tools were called, with which arguments, in order (allow extra reads) over fixed seed cases; hold writes to an exact count. This survives non-determinism that exact-match scoring cannot.
  2. Grow the set from real failures. When a trace shows a wrong or unsafe trajectory, distil it into one case that pins that single behavior and makes a recurrence visible.
  3. Add a groundedness check. Require every recognized claim to appear in that turn's retrieved evidence or the user's question. Document the recognizer's vocabulary; use a broader extractor or judge for claims it cannot parse.
  4. Warn on token drift. Total this run's tokens and model calls, compare them with the previous run of the same evalset and model, and print the change when it exceeds a stated tolerance — 25% in the reference implementation. Keep it a warning, not a gate: tokens move for honest reasons, and a run that answered every case correctly should not fail for spending more to do it. Trajectory scores tolerate waste, so this is the only signal that surfaces a correct-but-expensive change at all.
  5. A/B prompt versions. Run the eval set under two pinned prompt versions in isolated processes and print a per-scorer delta; promote or roll back on the numbers.
  6. Split gates from evidence. Deterministic, model-free checks gate CI; model-backed evaluations run as commit-scoped evidence a human reads before release, with the thresholds visible on the command that produced them.

Read the full file on GitHub · 37 lines

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. 3d ago First seen · 37 lines · 71 tokens per session scan A 1539d25490ca

Subscribe to this mod's changes

agent-evaluation is a skill published in the GitHub repository MLOps-Courses/agentops-open-course (2 stars, last pushed 10d ago), licensed MIT. It adds 71 tokens to every session and 728 once invoked, about $0.0004 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 2026-08-31.

Related

Other skills, from other repositories

find-your-level

Interactive quiz that maps your AI/ML knowledge to a starting point in the 523-lesson, 20-phase AI Engineering from Scratch curriculum. Trigger phrases: "where should I start", "find my level", "what do I know", "which phase", "assess my knowledge", "placement test", "skip ahead".

rohitg00/ai-engineering-from-scratch · 71 tokens

skill-release-gate

Evaluate an Agent Skill bundle for structural integrity, trigger quality, artifact improvement, script correctness, safety, installed-tree integrity, and target-host portability before release.

rohitg00/ai-engineering-from-scratch · 36 tokens

migration-review

Review database migration files when a change adds or modifies paths under migrations/. Use it before merge to collect forward, rollback, locking, and data-safety evidence.

rohitg00/ai-engineering-from-scratch · 35 tokens

command-creator

This skill should be used when creating a Claude Code slash command. Use when users ask to "create a command", "make a slash command", "add a command", or want to document a workflow as a reusable command. Essential for creating optimized, agent-executable slash commands with proper structure and best practices.

meshery/meshery · 66 tokens

gen-test

Generate idiomatic tests for Go packages and handlers in the Meshery project.

meshery/meshery · 18 tokens

superplane-changelog

When generating a SuperPlane changelog from merged commits. Use for "what's new" summaries with new integrations, new components/triggers, improvements, security updates, and bug fixes. Output is user-focused markdown in tmp/.

superplanehq/superplane · 50 tokens