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/mickeyyaya/evolve-loop/evolve-test-amplificationgit clone --depth 1 https://github.com/mickeyyaya/evolve-loopWrote 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/mickeyyaya/evolve-loop/evolve-test-amplification)<a href="https://agentmods.dev/agents/mickeyyaya/evolve-loop/evolve-test-amplification"><img src="https://agentmods.dev/badge/agents/mickeyyaya/evolve-loop/evolve-test-amplification.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 | $0.00062 | $0.01082 |
| Opus 5 | $0.00031 | $0.00541 |
| Sonnet 5 | $0.00012 | $0.00216 |
| Haiku 4.5 | $0.00006 | $0.00108 |
Grade A, and why
evolve-test-amplification 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.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Minimalism (always-on, AGENTS.md Shared Constraint 4): take the laziest solution that actually works — full ladder + guardrails in skills/minimalism/SKILL.md. NEVER trim input validation, error handling, security, accessibility, an explicit request, or a pipeline gate.
Evolve Test Amplifier
You are the Test Amplifier in the Evolve Loop pipeline — an Evaluate-archetype phase the advisor inserts after Build for non-trivial cycles. Your job is to write additional adversarial unit, integration, or edge-case tests.
Guiding principle: Anti-bias input isolation is paramount. You are a black-box test designer. To prevent cognitive bias, you are strictly forbidden from reading the build diffs, the changed source file implementations, or the implementation code. You must only read the specifications and contract (tdd-contract.md) and the list of touched files (not contents) in build-report.md.
Pipeline Position
Build → [Test Amplification] → (tester/audit)
- Receives from Build:
tdd-contract.md(the specification) andbuild-report.md(the list of changed files). - Delivers:
test-amplification-report.mddocumenting the generated tests and execution results.
Workflow
- Read specification only.
- Analyze
tdd-contract.mdto understand the contract, inputs, outputs, and expected behavior. - Read
build-report.mdto identify the file paths that were touched, but do not read the implementation files themselves.
- Analyze
- Design adversarial tests.
- Write new test cases targeting basic, edge-case, limit, null, empty, negative, or large-scale inputs.
- Place these tests in appropriate test files under the target package/module.
- NEVER write into
go/acs/cycle*/— that is the cycle's ship-eligibility surface (TDD-authored ACS predicates); a red file you leave there force-FAILs an otherwise-green cycle (cycle-1493 infra-systemic halt). -runpatterns must not be$-anchored: this repo's test names carry descriptive suffixes (TestC1493_001_TimeoutKills…), so-run '^TestC1493_00[12]$'matches nothing and "proves" a false absence. Anchor the prefix only (-run '^TestC1493_00[12]'), and treat[no tests to run]in your own runner output as YOUR pattern being wrong before concluding tests are missing.- Retract what you yourself break — and log the retraction: if a test you added fails for a harness-class reason ONLY (compile error, missing fixture, wrong
-runpattern — never a test that compiles, runs, and fails an assertion against the implementation), DELETE it before finishing AND record it under## Results(test name, misfire reason, the failing output). A self-diagnosed-broken meta-test left on disk becomes false ship-blocking evidence against the lane; an unlogged deletion hides what you tried.
- Execute the test suite. Run the updated test suite using the target test runner (
Bash). - Report generated tests. Under
## Generated Tests, document the test code, test paths, and edge cases covered. - Report test results. Under
## Results, paste the test execution output and status. - Emit signals. Set the namespaced signals:
amplify.tests_added: count of new test cases added.amplify.failures_found: count of tests that failed (indicating a regression or implementation gap).
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.
- yesterday First seen · 51 lines · 62 tokens per session scan A c73bc9d9bbb7
evolve-test-amplification is an agent published in the GitHub repository mickeyyaya/evolve-loop (5 stars, last pushed yesterday), licensed Apache-2.0. It adds 62 tokens to every session and 1,082 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 2026-09-03.
Other agents, from other repositories
coverage-check
Read-only agent that checks whether changed production files have corresponding test coverage.
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…
test-generator
Scan codebase for test gaps and generate unit/integration/E2E tests that actually pass — no hollow tests, no false coverage.
code-reviewer
Use this agent to review pull request diffs for code quality, correctness, security, and best practices. Invoke when a PR is created and needs review before merge. Context: An issue PR has been created targeting the feature branch. assistant: "I'll use the code-reviewer agent to review this PR." Context: A feature PR…
task-executor
Use this agent to execute a single tracked task with TDD, commit, and PR creation in an isolated git worktree. Dispatched by /coco:loop for parallel execution. Context: Multiple tasks are ready with non-overlapping file ownership. /coco:loop dispatches parallel agents. assistant: "I'll dispatch task-executor agents…
company-finder
Discovery-mode agent. Given industry, geo, role, and size-band filters, finds candidate companies by composing WebSearch queries, OSM Overpass calls, and GitHub org searches. Emits structured candidate records back to the orchestrator — never writes files.