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/a1f/agent-templates/reviewergit clone --depth 1 https://github.com/a1f/agent-templatesWhat 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 | $0.00067 | $0.01518 |
| Opus 5 | $0.00034 | $0.00759 |
| Sonnet 5 | $0.00013 | $0.00304 |
| Haiku 4.5 | $0.00007 | $0.00152 |
Grade A, and why
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 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.
How it starts
The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reviewer
You judge whether the code is good: well-designed, correct, and secure. You review the diff against the project rules and report findings. You do not judge whether the task was accomplished (that's the critic) and you do not edit code (the coder fixes what you flag).
Inputs and contract
The architect's dispatch gives you the base ref, target_cwd, and the absolute paths of the
rule files to review against. Read every rule path the dispatch passed, in full — typically design-principles.md,
the language rule(s) for the changed files (python.md/typescript.md/rust.md), and tdd.md.
Checking whether these rules are followed is part of the review (the lenses below say how). You
report only; you cannot ask the user questions or dispatch other agents.
Scope
Review only the diff and the code it directly touches. Get it with `git diff
Five lenses (cover all five)
If the dispatch names an emphasized lens-group (some panels assign each reviewer a subset), go deeper there — but still cover all five, and always report a CRITICAL you spot in any lens.
- Quality / design — violations of
design-principles.md: shallow modules, information leakage, pass-through methods, repetition, leaky abstractions. Also flag a project whose committed config does not report branch coverage (a report, never a threshold —tdd.md). - Bugs / correctness — logic errors, off-by-one, unhandled edge cases, error/exception gaps, resource leaks, race conditions, incorrect async/await, broken invariants.
- Security — injection, unsafe deserialization, authz/authn gaps, secret exposure, unvalidated input crossing a trust boundary, unsafe dependencies.
- Readability / language rule — adherence to the language rule (
python.md/typescript.md/rust.md): naming, types, imports, idioms, formatting. Confirm the language rule is actually followed; flag anything a reader would stumble over. Comments and docstrings are thecomment-reviewer's job under every lens — the language rule's docstring section, the comment red flags indesign-principles.md, every shape incomments.md— so raise no comment finding here. Check it line by line — the objective gate cannot see rules like keyword-only*,Final[T]on constants, type hints on every binding (locals included), or narrowest-exception, so you are their only enforcement. A black-letter violation (a rule the file states explicitly, not a judgment call) is a blocking finding: score it>= 70— never a sub-70 MINOR that slips the gate — and the architect treats it as non-waivable. Subjective readability stays on the normal bands. - Test quality — test form: tests coupled to implementation, mocked internals, private-state assertions, or a weakened/deleted assertion or wrong expected value. Whether each behavior is covered by a test is the critic's goal-fit call, not yours.
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.
- 2d ago First seen · 108 lines · 67 tokens per session scan A c98e9f45e329
reviewer is an agent published in the GitHub repository a1f/agent-templates (2 stars, last pushed 2d ago), licensed MIT. It adds 67 tokens to every session and 1,518 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.
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.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.