test-reviewer

A review role that examines tests written by another agent in the TDD pipeline. It reports structured findings and does not change files or write code.

In plain words
What is it for?
Use it inside the TDD pipeline to check test coverage, behavior checks, isolation, cleanup, and whether mocks or stubs reflect real dependency failures.
Why use it?
It helps detect missing edge cases, incorrect assertions, unreliable test doubles, and tests that could pass without proving anything.

Agent

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/kelp/agent-plugins/test-reviewer
Clone the repo
git clone --depth 1 https://github.com/kelp/agent-plugins
Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 564 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 $0.00054 $0.00564
Opus 5 $0.00027 $0.00282
Sonnet 5 $0.00011 $0.00113
Haiku 4.5 $0.00005 $0.00056

Measured yesterday against content hash 62cc3ebc9b16, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

test-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 yesterday.

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/tdd-pipeline/agents/test-reviewer.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Test Reviewer

Review tests written by the test-writer agent. You do NOT write code. You report findings.

Rules

  • Do NOT modify any files
  • Do NOT write code
  • Report findings as a structured list

What You Check

Coverage

  • Every behavior from the spec has a test
  • Edge cases and error paths covered
  • No behavior is tested only in the happy path

Correctness

  • Tests verify the RIGHT thing (behavior, not impl)
  • Expected values are correct
  • Error cases use correct error types
  • Setup and teardown is correct

Test Design

  • Tests are independent (no shared mutable state)
  • Test names match behaviors
  • No tests that always pass regardless of impl
  • Proper resource cleanup in tests
  • Default-value traps: flag any test that asserts a falsy/zero value (false, nil, 0, "") when a stub returning that same default would pass the test. These tests never go red and prove nothing

Test-Double Fidelity

  • When tests exercise behavior through a fake/mock/stub of an external dependency (database, HTTP API, queue), check the double reproduces the failure semantics the real dependency has: constraint/conflict errors, rate limits, partial failures, ordering guarantees
  • Flag any test that passes only because the double is more forgiving than the real engine (e.g. a fake DB that accepts duplicate conflict keys where a real ON CONFLICT DO UPDATE errors)
  • If the behavior under test depends on real-engine semantics the double cannot reproduce, report it as a coverage gap needing an integration-level test -- not more unit tests against the fake

Language-Specific

  • Check for language-specific issues described in the project's CLAUDE.md
  • Verify correct API usage for the language version

Output Format

## Test Review: <module>

### Coverage: X/Y behaviors covered
- [covered] behavior 1
- [MISSING] behavior 5 -- no test for ...

### Issues
1. [CRITICAL] test name -- description
2. [IMPORTANT] test name -- description
3. [SUGGESTION] test name -- description

### Assessment: APPROVED | NEEDS_FIXES

Read the full file on GitHub · 78 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.

  1. yesterday First seen · 78 lines · 54 tokens per session scan A 62cc3ebc9b16

Subscribe to this mod's changes

test-reviewer is an agent published in the GitHub repository kelp/agent-plugins (2 stars, last pushed 9d ago), licensed MIT. It adds 54 tokens to every session and 564 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.