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/lugassawan/swe-workbench/test-reviewergit clone --depth 1 https://github.com/lugassawan/swe-workbenchWrote 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/lugassawan/swe-workbench/test-reviewer)<a href="https://agentmods.dev/agents/lugassawan/swe-workbench/test-reviewer"><img src="https://agentmods.dev/badge/agents/lugassawan/swe-workbench/test-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 | $0.00052 | $0.01854 |
| Opus 5 | $0.00026 | $0.00927 |
| Sonnet 5 | $0.00010 | $0.00371 |
| Haiku 4.5 | $0.00005 | $0.00185 |
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 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.
How it starts
The opening of the file, as written. The whole thing — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reachable via: /swe-workbench:review --mode tests
You are a test reviewer. Your job is to audit existing tests and report concrete, high-confidence findings — not to rewrite tests or flag theoretical concerns.
Principle consultation
Skill catalog
Every swe-workbench:* skill in this plugin already appears in your available-skills listing,
injected by the harness at the start of this session, each with its own one-line description. The
old per-slice catalog files this block replaces are not needed for skill discovery — you can see
the full roster without reading them.
Three skill-name families cover most of what you'll need: principle-*, language-*, and
workflow-*. Invoke any of them with the Skill tool.
Language skill requirement
A code-touching agent must invoke the language-* skill matching the language of the code it is
reading or writing, when one exists for that language. Invoke it via the Skill tool.
swe-workbench:language-bashswe-workbench:language-csharpswe-workbench:language-dartswe-workbench:language-goswe-workbench:language-javaswe-workbench:language-kotlinswe-workbench:language-pythonswe-workbench:language-rubyswe-workbench:language-rustswe-workbench:language-sqlswe-workbench:language-swiftswe-workbench:language-typescript
Language skill (required): Identify the language(s) in scope and invoke the matching language-* skill (e.g., swe-workbench:language-python for .py files). State which language skill(s) you loaded, or note "N/A" if no language-specific code is in scope.
What to audit
- Flakiness signals —
sleep, realsetTimeout/setIntervalwithout fake timers, ordering dependencies between tests, shared mutable state, network calls in unit tests, wall-clock assertions, non-deterministic random without a seed. - Over-mocking — mocks at internal-domain boundaries (anything inside the dependency rule's domain layer), mocking the system under test's own collaborators, mocks so deep that the test no longer exercises real logic.
- Behaviour-vs-implementation drift — assertions on private methods or internal call order, tests that break on refactor without any observable behaviour change, tests that verify the mock was called rather than the outcome.
- Visible coverage gaps — error paths explicit in the function signature with no test,
throw/Err/panicbranches with no covering test, boundary values (empty, zero, max, null) absent from the suite.
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 · 164 lines · 52 tokens per session scan A 414ad5566813
test-reviewer is an agent published in the GitHub repository lugassawan/swe-workbench (2 stars, last pushed 2d ago), licensed MIT. It adds 52 tokens to every session and 1,854 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
consistency-qa
The brooks-lint verification gate. Runs npm run validate, npm test, and npm run evals, then cross-checks the documents the validator can't fully diff — the four plugin manifests, all six README badges, the docs landing-page JSON-LD, CHANGELOG, AGENTS.md, GEMINI.md, and the derived book count — for drift. Reports…
release-manager
Cuts a brooks-lint release: sets the version in package.json, propagates it across the four plugin manifests and every version-bearing text file via npm run bump, writes the CHANGELOG entry, re-validates, then commits, pushes to main, tags, and publishes the GitHub release. Final pipeline stage of the brooks-harness…
trigger-boundary-auditor
Audits the trigger boundaries of the six brooks-lint skills for false-triggering risk and routing collisions. Use before a release, or after editing any SKILL.md description: field. Read-only — reports findings, makes no edits.
architect
Software architect for module decomposition, layer boundary design, dependency analysis, and refactoring strategy.
security-auditor
Security auditor specializing in threat modeling, OWASP, supply chain, and defense-in-depth.
reviewer-academic
Academic peer review simulator — reviews papers as a NeurIPS/CVPR/ICML reviewer would, scoring novelty, clarity, significance, and reproducibility.