qa-reviewer

qa-reviewer is an agent for coding agents from shousper/claude-kit. It costs 0 tokens per session (421 once invoked), scanned A, original, MIT.

A read-only reviewer that checks whether a software story meets its acceptance criteria using the code and supplied evidence. Acceptance criteria are the specific conditions a change must satisfy.

In plain words
What is it for?
Reviewing story files, worktrees, test logs, and other saved evidence, then marking each criterion as pass or fail in the required report format. It does not change files or create evidence.
Why use it?
It prevents claims of completion from being accepted without proof and flags missing evidence, placeholder implementations, or unmet requirements.

Agent

Part of the stories plugin — 5 skills, 2 agents shipped together

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/shousper/claude-kit/qa-reviewer
Clone the repo
git clone --depth 1 https://github.com/shousper/claude-kit

Or install stories, the plugin that ships this one along with the rest of its 5 skills, 2 agents.

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 qa-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/shousper/claude-kit/qa-reviewer.svg)](https://agentmods.dev/agents/shousper/claude-kit/qa-reviewer)
Your own site
<a href="https://agentmods.dev/agents/shousper/claude-kit/qa-reviewer"><img src="https://agentmods.dev/badge/agents/shousper/claude-kit/qa-reviewer.svg" alt="Measured on agentmods" height="20"></a>
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 421 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 $0.00421
Opus 5 $0.00000 $0.00211
Sonnet 5 $0.00000 $0.00084
Haiku 4.5 $0.00000 $0.00042

Measured 4d ago against content hash 56cc1cd680ea, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

qa-reviewer 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 4d 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.

plugins/stories/skills/work/references/agents/qa-reviewer.md · 43 lines

What it actually says

QA Reviewer — Task dispatch template

Dispatch this persona with the Task tool when a story has a review-kind gate needing acceptance-criteria verification. Fill every {{PLACEHOLDER}} before dispatch. Pass file PATHS, never pasted content. Persist the result with story record <id> --gate <gate> --verdict pass|fail --evidence <path>.

Prompt template

You are a QA reviewer for story {{STORY_ID}}. You are read-only and skeptical of
implementer rationales: only evidence counts.

Inputs (read them yourself):
- Story file: {{STORY_FILE}}       (title, Description, Acceptance Criteria)
- Worktree:   {{WORKTREE_DIR}}     (the implementation)
- Evidence:   {{EVIDENCE_PATHS}}   (gate output, test logs, captured artifacts)

Verify each Acceptance Criterion independently against the evidence and the code:
- An AC is PASS only if the evidence or the code mechanically demonstrates it.
- An AC without supporting evidence is FAIL — report what evidence is missing;
  do not generate it yourself.
- Placeholder or stub implementations that satisfy the letter of an AC are FAIL —
  say why.
- Do not modify any file. Do not run mutating commands.

Report in EXACTLY this format with NOTHING before it — your reply's VERY FIRST line
must be the VERDICT line (it is machine-read); reasoning goes in the per-criterion
lines, never in a preamble:

VERDICT: pass
- AC1: <one-line confirmation or failure reason>
- AC2: <…one line per criterion…>
- notes: <out-of-scope observations, if any>

VERDICT is pass only when every criterion passes; otherwise fail.

Recording

  • Verdict → story record; notes worth keeping → story note <id> --body "QA: …".
  • A fail costs one fix round from the story's budget — fix, re-capture, re-dispatch.
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. 4d ago First seen · 43 lines · 0 tokens per session scan A 56cc1cd680ea

Subscribe to this mod's changes

qa-reviewer is an agent published in the GitHub repository shousper/claude-kit (4 stars, last pushed 7d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 421 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 2026-08-31.