askit-quality-grader

askit-quality-grader is an agent for Claude Code from product-on-purpose/agent-skills-toolkit. It costs 60 tokens per session (537 once invoked), scanned A, original, Apache-2.0.

A delegated role that tests a skill against example cases and judges whether it triggers and behaves as expected.

In plain words
What is it for?
Use it for behavioral evaluation of skills, including checking trigger cases and expected outputs without changing the skill being tested.
Why use it?
It helps reveal when a skill fires for the wrong requests or produces behavior that does not match its expected results.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the agent-skills-toolkit plugin — 27 skills, 2 commands, 7 agents, 1 hook, 1 plugin shipped together

Good fit Use it for behavioral evaluation of skills, including checking trigger cases and expected outputs without changing the skill being tested.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/product-on-purpose/agent-skills-toolkit/askit-quality-grader
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.

Clone the repo
git clone --depth 1 https://github.com/product-on-purpose/agent-skills-toolkit

Made for: Claude Code.

Or install agent-skills-toolkit, the plugin that ships this one along with the rest of its 27 skills, 2 commands, 7 agents, 1 hook, 1 plugin.

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 askit-quality-grader

README.md
[![agentmods](https://agentmods.dev/badge/agents/product-on-purpose/agent-skills-toolkit/askit-quality-grader/github.svg)](https://agentmods.dev/agents/product-on-purpose/agent-skills-toolkit/askit-quality-grader)
Your own site
<a href="https://agentmods.dev/agents/product-on-purpose/agent-skills-toolkit/askit-quality-grader"><img src="https://agentmods.dev/badge/agents/product-on-purpose/agent-skills-toolkit/askit-quality-grader/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 askit-quality-grader

Your own site · 80×15
<a href="https://agentmods.dev/agents/product-on-purpose/agent-skills-toolkit/askit-quality-grader"><img src="https://agentmods.dev/badge/agents/product-on-purpose/agent-skills-toolkit/askit-quality-grader.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 537 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 $0.00060 $0.00537
Opus 5 $0.00030 $0.00269
Sonnet 5 $0.00012 $0.00107
Haiku 4.5 $0.00006 $0.00054

Measured 9d ago against content hash 7ee936262ae7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

askit-quality-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 9d 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.

agents/askit-quality-grader.md · 27 lines

What it actually says

askit-quality-grader

Role

The behavioral-judge delegate behind askit-evaluate's behavioral mode. Runs a skill against its eval-set (the triggering {query, should_trigger} cases and the {given, expect} behavior cases under evals/) and judges, case by case, whether the skill fires when it should, stays silent when it should not, and produces the expected behavior. It reports a verdict per case with evidence. This is the LLM-judged layer the Standard defers as roadmap (the multi-tier eval engine, ADR 0023); it produces evidence beside the deterministic gate and never returns a CI pass/fail (Design Principle 3). It is distinct from askit-evaluator (deterministic conformance) and askit-reviewer (qualitative review of the artifact, not its runtime behavior).

Tools

Read to load the skill and its eval-set; Bash to exercise the skill and the harness as needed (Standard sec 9, narrowest set). No write access (judging must not mutate what it grades).

Steps

  1. Read the target skill and its evals/ cases (triggering and behavior). If no evals/ exists (the common case - the convention is forward-looking), derive the case set: should-fire queries and adversarial near-miss no-fire queries from the description (read sibling skills to make the near-misses competitive), and behavior cases from the documented workflow. Note the derivation in the evidence.
  2. For each case, exercise the skill and judge fire / no-fire and the output against the expectation; when live execution is not possible, judge by static analysis of the artifact and say so.
  3. Report a per-case verdict (pass / fail) with the evidence and a short reason, then summarize the pass rate (fired = should-fire cases that fire, missed = should-fire cases that do not; a false fire is a failed case). The result is advisory evidence, not a gate result.
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. 9d ago First seen · 27 lines · 60 tokens per session scan A 7ee936262ae7

Subscribe to this mod's changes

askit-quality-grader is an agent published in the GitHub repository product-on-purpose/agent-skills-toolkit (2 stars, last pushed 3d ago), licensed Apache-2.0. It adds 60 tokens to every session and 537 once invoked, about $0.0003 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.