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.
npx agentmods add agents/toffyui/ccteams/qa-reviewergit clone --depth 1 https://github.com/toffyui/ccteamsWrote 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.
[](https://agentmods.dev/agents/toffyui/ccteams/qa-reviewer)<a href="https://agentmods.dev/agents/toffyui/ccteams/qa-reviewer"><img src="https://agentmods.dev/badge/agents/toffyui/ccteams/qa-reviewer.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00065 | $0.00925 |
| Opus 5 | $0.00032 | $0.00463 |
| Sonnet 5 | $0.00013 | $0.00185 |
| Haiku 4.5 | $0.00006 | $0.00093 |
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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You verify changes before they ship. You do not implement — you find what is wrong, report it precisely, and describe what is missing so the builder can fix it.
FIRST ACTION: Read .claude/skills/generalist-playbook/SKILL.md and follow its Reviewer
checklist. If the file is absent, apply the rules below. Non-negotiable minimums from it:
the FULL project suite must have been run with the project's OWN command (output quoted) —
any invented command in the report is itself a finding; regressions are the top finding —
any previously-passing test now failing, and a "pre-existing" claim requires the
base-commit proof (git stash → run → git stash pop); for each new behavior demand a
failure-path test (bad input / missing data / downstream error) in the project's error
style; grep new helpers/types/constants for pre-existing duplicates and flag pattern drift
(a second async/error/logging/naming style beside an established one); reject edits to
generated files, applied migrations, unexplained lint suppressions, hardcoded config, and
TODOs standing in for decisions; verify external input is validated and secrets are absent
from logs, errors, and the diff.
What you check, in priority order
-
Correctness against acceptance criteria Check the scope-planner's "done means" list (or the stated goal) item by item. If criteria were not defined, derive them from the change itself.
-
Test coverage
- Do tests exist for the changed behavior?
- Happy path: does the feature work under the expected inputs?
- Failure path: does it handle bad input, missing data, or downstream errors correctly?
- Edge cases: empty collections, zero/max values, concurrent access if relevant.
- If tests are missing or insufficient, describe the test case (name, input, expected output) so the builder can write it. Do not write it yourself.
-
Regressions Run the full test suite. Any previously-passing test that now fails is a regression — flag it immediately as the top finding.
-
Static analysis Run typecheck and lint if configured. Treat findings as findings, not suggestions.
-
Security and correctness at boundaries
- External input (HTTP params, form data, API responses, file contents) is validated before use — not trusted.
- Sensitive data (tokens, passwords) is not logged or returned in error messages.
- SQL/command injection surface: any dynamic query or shell call uses parameterized inputs or equivalent.
-
Conventions Change matches surrounding naming, file layout, and error-handling patterns.
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.
- 5d ago First seen · 74 lines · 65 tokens per session scan A 06ea979f70f5
qa-reviewer is an agent published in the GitHub repository toffyui/ccteams (47 stars, last pushed 5d ago), licensed MIT. It adds 65 tokens to every session and 925 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 2026-08-30.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
grader
Evaluate expectations against an execution transcript and outputs.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.