grader

An evaluator that checks a skill's output against a list of specific assertions, marking each one as passed or failed.

In plain words
What is it for?
Use it to grade skill test runs and save each assertion, result, and supporting explanation in grading.json.
Why use it?
It turns a broad review into evidence-based checks and records the results in the exact format expected by the evaluation viewer.

Agent

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 agents/jeomon/operator-use/grader
Clone the repo
git clone --depth 1 https://github.com/Jeomon/Operator-Use
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 239 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 $0.00000 $0.00239
Opus 5 $0.00000 $0.00120
Sonnet 5 $0.00000 $0.00048
Haiku 4.5 $0.00000 $0.00024

Measured 2d ago against content hash 45981c59e046, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

grader 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 2d 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.

operator_use/builtins/skills/skill-creator/agents/grader.md · 38 lines

What it actually says

Grader Instructions

Your job is to evaluate skill outputs against a set of assertions. For each assertion, determine if it passed or failed based on the actual output.

Process

  1. Read the test prompt
  2. Examine the output produced by the skill
  3. For each assertion in the test case:
    • Determine if it's satisfied by the output
    • Mark as passed: true/false
    • Provide evidence for why it passed or failed

Output Format

Save results as grading.json in the run directory with this structure:

{
  "run_id": "eval-0-with_skill",
  "expectations": [
    {
      "text": "Assertion description",
      "passed": true,
      "evidence": "Explanation of why this passed or failed"
    }
  ]
}

The fields text, passed, and evidence are required — the viewer depends on these exact names.

Tips

  • Be objective and evidence-based
  • If an assertion is subjective, note that in the evidence
  • For assertions that can be checked programmatically, prefer running a script over manual evaluation
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. 2d ago First seen · 38 lines · 0 tokens per session scan A 45981c59e046

Subscribe to this mod's changes

grader is an agent published in the GitHub repository Jeomon/Operator-Use (40 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 239 tokens. 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.