agent-evaluation

agent-evaluation is a skill for Claude Code from autohandai/community-skills. It costs 49 tokens per session (3,361 once invoked), scanned A, original, Apache-2.0.

A guide to evaluating AI agents, meaning systems that take actions to complete tasks, with tests, scoring methods, and monitoring.

In plain words
What is it for?
It helps build benchmarks and test cases for coding, conversation, research, or computer-use agents; create automated or human graders; and connect evaluations to production monitoring and CI/CD.
Why use it?
It helps measure whether an agent works reliably and whether changes improve its results over time.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Good fit It helps build benchmarks and test cases for coding, conversation, research, or computer-use agents; create automated or human graders; and connect evaluations to production monitoring and CI/CD.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/autohandai/community-skills/agent-evaluation
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 autohandai/community-skills --skill agent-evaluation
Clone the repo
git clone --depth 1 https://github.com/autohandai/community-skills

Made for: Claude Code.

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/autohandai/community-skills/agent-evaluation/github.svg)](https://agentmods.dev/skills/autohandai/community-skills/agent-evaluation)
Your own site
<a href="https://agentmods.dev/skills/autohandai/community-skills/agent-evaluation"><img src="https://agentmods.dev/badge/skills/autohandai/community-skills/agent-evaluation/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agent-evaluation

Your own site · 80×15
<a href="https://agentmods.dev/skills/autohandai/community-skills/agent-evaluation"><img src="https://agentmods.dev/badge/skills/autohandai/community-skills/agent-evaluation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,361 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00049 $0.03361
Opus 5 $0.00024 $0.01681
Sonnet 5 $0.00010 $0.00672
Haiku 4.5 $0.00005 $0.00336

Measured 10d ago against content hash 4c98974f8938, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

agent-evaluation scanned grade A with 1 finding 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 10d 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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

result = subprocess.run(
agent-evaluation/SKILL.md · 484 lines

How it starts

The opening of the file, as written. The whole thing — 484 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Evaluation (AI Agent Evals)

Based on Anthropic's "Demystifying evals for AI agents"

When to use this skill

  • Designing evaluation systems for AI agents
  • Building benchmarks for coding, conversational, or research agents
  • Creating graders (code-based, model-based, human)
  • Implementing production monitoring for AI systems
  • Setting up CI/CD pipelines with automated evals
  • Debugging agent performance issues
  • Measuring agent improvement over time

Core Concepts

Eval Evolution: Single-turn → Multi-turn → Agentic

Type Turns State Grading Complexity
Single-turn 1 None Simple Low
Multi-turn N Conversation Per-turn Medium
Agentic N World + History Outcome High

7 Key Terms

Term Definition
Task Single test case (prompt + expected outcome)
Trial One agent run on a task
Grader Scoring function (code/model/human)
Transcript Full record of agent actions
Outcome Final state for grading
Harness Infrastructure running evals
Suite Collection of related tasks

Instructions

Step 1: Understand Grader Types

Code-based Graders (Recommended for Coding Agents)
  • Pros: Fast, objective, reproducible
  • Cons: Requires clear success criteria
  • Best for: Coding agents, structured outputs
# Example: Code-based grader
def grade_task(outcome: dict) -> float:
    """Grade coding task by test passage."""
    tests_passed = outcome.get("tests_passed", 0)
    total_tests = outcome.get("total_tests", 1)
    return tests_passed / total_tests

# SWE-bench style grader
def grade_swe_bench(repo_path: str, test_spec: dict) -> bool:
    """Run tests and check if patch resolves issue."""
    result = subprocess.run(
        ["pytest", test_spec["test_file"]],
        cwd=repo_path,
        capture_output=True
    )
    return result.returncode == 0

Read the full file on GitHub · 484 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. 10d ago First seen · 484 lines · 49 tokens per session scan A 4c98974f8938

Subscribe to this mod's changes

agent-evaluation is a skill published in the GitHub repository autohandai/community-skills (11 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 49 tokens to every session and 3,361 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). 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

dataset-evaluation

Validates dataset formatting and quality for SageMaker model fine-tuning (SFT, DPO, or RLVR). Use when the user says "is my dataset okay", "evaluate my data", "check my training data", "I have my own data", or before starting any fine-tuning job. Detects file format, checks schema compliance against the selected model…

awslabs/agent-plugins · 93 tokens

agentsop-domain-eval-set

Build and govern a 50-200 example domain-specific held-out benchmark sampled from real traffic. Distinct from public benchmarks (MMLU/HumanEval/GSM8K via lm-evaluation-harness) which measure GENERAL capability. Only a held-out domain set predicts whether THIS system works on YOUR data. Collect real examples, label…

agentsope/SkillAlchemy · 98 tokens

agentsop-regression-gate

Build a held-out eval set, run it on every prompt/model change, and block regressions in CI. An LM change is a code change — gate it with a test suite (eval set + metric + threshold). Cross-framework SOP not surfaced by any single base skill.

agentsope/SkillAlchemy · 62 tokens

agentsop-test-fix-loop

Decision protocol for wiring a verify-then-fix loop around a code-editing LLM agent. The agent edits → runs lint/test → reads the output → fixes → re-runs, bounded by an iteration cap and an escalation rule. Activates whenever a coder agent has a verifiable success criterion (exit code, type-checker output, failing…

agentsope/SkillAlchemy · 146 tokens

test-driven-development

TDD: enforce RED-GREEN-REFACTOR, tests before code.

HezaoHezao/poirot · 20 tokens

test-roadmap

EXPERIMENTAL. Analyzes a repository and any existing test suite, grades existing tests for weakness, classifies mocks, emits a phased roadmap for building a test suite that catches real regressions, then executes those phases one at a time. Use when planning or building a test suite, assessing whether existing tests…

Ovid/paad · 111 tokens