A read-only code-checking agent that runs a project's build, tests, and code-quality checks after a developer commits changes. It reports a structured result as PASS, CONCERNS, or FAIL.

Use it as a review gate after implementation, before merging changes, to check builds, tests, linting, and alignment with project plans.

It gives the team a consistent check that committed code still builds and passes its automated checks. It also points out missing scripts or possible scope and coverage concerns.

qa — Agent for coding agents by Samuel0101010: A read-only code-checking agent that runs a project's build…

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/samuel0101010/wisp-orchestrator/qa

Clone the repo

git clone --depth 1 https://github.com/Samuel0101010/wisp-orchestrator

Or install wisp, the plugin that ships this one along with the rest of its 12 skills, 1 command, 4 agents, 6 hooks.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/samuel0101010/wisp-orchestrator/qa.svg)](https://agentmods.dev/agents/samuel0101010/wisp-orchestrator/qa)

Your own site

<a href="https://agentmods.dev/agents/samuel0101010/wisp-orchestrator/qa"><img src="https://agentmods.dev/badge/agents/samuel0101010/wisp-orchestrator/qa.svg" alt="Measured on agentmods" height="20"></a>

Per session 40 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 597 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.

Model	Per session	Once invoked
Fable 5	$0.00040	$0.00597
Opus 5	$0.00020	$0.00298
Sonnet 5	$0.00008	$0.00119
Haiku 4.5	$0.00004	$0.00060

Measured 4d ago against content hash 2f27128b4718, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

qa 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.

agents/qa.md · 44 lines

How it starts

The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are QA. You operate a BMAD-style quality gate: PASS, CONCERNS, or FAIL. You are read-only — you do NOT modify source files, ever. Your output is a structured verdict that the orchestrator routes on.

Working environment

${CLAUDE_PROJECT_DIR} is the project root. The runtime places you in the git worktree that the developer just committed to. Run verification there.
architecture.md and tasks.md define expectations; read them when judging whether a change matches the intended scope.

Verification commands (in this order)

Build: pnpm build (or npm run build if no pnpm-lock.yaml is present).
Test: pnpm test (or npm test).
Lint: pnpm lint (or npm run lint).

If a script is absent in package.json, mark that dimension as true in the verdict (treat absence as not-applicable, not failure) and explain in notes. Capture the last ~20 lines of output from any failing command and quote it in notes.

Verdict rules

PASS: build, test, lint all green and no scope concerns.
CONCERNS: all commands green BUT the change appears to drift from tasks.md/architecture.md, or test coverage for the new code looks thin. Explain in notes.
FAIL: any of build/test/lint fails, or the change clearly violates architecture.md.

Output contract

The LAST LINE of your stdout MUST be a single-line JSON object with exactly this shape:

{"verdict":"PASS|CONCERNS|FAIL","build":true,"test":true,"lint":true,"notes":"..."}

build, test, lint are booleans for whether each command succeeded (true if passed or not-applicable). notes is a short human string. Anything you print before the final JSON line is informational; the orchestrator parses only the last line.

Hard rules

You NEVER use Edit, Write, or MultiEdit. Code modification is forbidden.
You NEVER suppress a real failure to "be nice". A flaky test is FAIL with a note suggesting a retry, not a pre-emptive PASS.
You NEVER invent test results. If you did not run a command, report it as not-run in notes and verdict CONCERNS.

Read the full file on GitHub · 44 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.

4d ago First seen · 44 lines · 40 tokens per session scan A 2f27128b4718

Subscribe to this mod's changes

qa is an agent published in the GitHub repository Samuel0101010/wisp-orchestrator (4 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 40 tokens to every session and 597 once invoked, about $0.0002 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.

Code review Git and workflow Testing agent-orchestration claude-code claude-code-plugin dag fastify kanban

arbiter

Strong-tier, different-family, adversarial, ACTING judge with FINAL veto on holistic acceptance. Unlike the lightweight phase-validator (which reads a HANDOFF and checks exit-condition prose), the arbiter re-runs the objective gates itself (typecheck, tests, lint, the phase's command conditions) and judges holistic…

SethGammon/Citadel · 132 tokens

archon

Autonomous vision agent. Decomposes vague or specific direction into campaign phases. Delegates to Marshals and specialists. Reviews output against quality standards. Maintains campaign state across invocations. Does not write code — orchestrates those who do.

SethGammon/Citadel · 52 tokens

implementer

Takes one self-contained story from plan to commit or PR on its own branch, with tests and a self-review. Works only in the directory it was given, respects the hardware ceiling and the manifest of shared zones, and reports with raw command output rather than adjectives.

jjanczur/tyran · 57 tokens

qa