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 commands/muhimu/claude-plan-lifecycle/bug-huntgit clone --depth 1 https://github.com/muhimu/claude-plan-lifecycleWrote 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/commands/muhimu/claude-plan-lifecycle/bug-hunt)<a href="https://agentmods.dev/commands/muhimu/claude-plan-lifecycle/bug-hunt"><img src="https://agentmods.dev/badge/commands/muhimu/claude-plan-lifecycle/bug-hunt.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 | $0.00013 | $0.00818 |
| Opus 5 | $0.00006 | $0.00409 |
| Sonnet 5 | $0.00003 | $0.00164 |
| Haiku 4.5 | $0.00001 | $0.00082 |
Grade A, and why
bug-hunt 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 — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Autonomous Bug Hunter
Act as an autonomous bug hunter for this issue:
$ARGUMENTS
Workflow — do not ask permission between steps
-
Hypothesize. Investigate the codebase with
Grep/Readto form 2–3 root-cause hypotheses. Write them as aTaskCreatelist (one task per hypothesis), ranked by likelihood with a one-line rationale each. -
Write the failing test. For the top hypothesis, write a test that reproduces the bug at the correct abstraction layer (unit if the bug is in a pure function, integration if it crosses package boundaries, end-to-end only if the bug is wiring/config). Prefer the smallest layer that can fail for the real reason.
-
Confirm red for the right reason. Run the test via
Bash. Read the failure output and verify it fails because of the hypothesized cause — not a compile error, missing import, wrong assertion, or unrelated panic. If the failure mode doesn't match the hypothesis, fix the test before touching production code. Mark the hypothesis task in-progress. -
Minimum fix. Implement the smallest change that turns the test green. No drive-by refactors, no extra error handling for cases the test doesn't cover, no speculative abstractions.
-
Verify green and clean. Run:
- the project's full test suite scoped to the affected module (e.g.
go test ./...,pytest) - the project's linter on the touched files Both must pass. If they don't, iterate — do not proceed.
- the project's full test suite scoped to the affected module (e.g.
-
Independent review. Spawn a reviewer subagent via the
Agenttool with a code-review subagent type if one is available (e.g.feature-dev:code-reviewer), elsegeneral-purpose. Give it:- the original symptom (verbatim from
$ARGUMENTS) - the diff (
git diff) - the new test
- an explicit instruction to check for regressions in adjacent code paths (callers of the changed function, sibling cases in the same switch/if-chain, related tests that were not run)
The reviewer must answer two questions: Does this fix the reported symptom? and What does this fix break?
- the original symptom (verbatim from
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 · 49 lines · 13 tokens per session scan A e81fff231fb0
bug-hunt is a command published in the GitHub repository muhimu/claude-plan-lifecycle (2 stars, last pushed 3d ago), licensed MIT. It adds 13 tokens to every session and 818 once invoked, about $0.0001 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 commands, from other repositories
tdd-requirements
TDD開発の要件整理を行います。機能要件を明確化し、テスト駆動開発のための準備を行います。.
pm-auto-design2dev
設計レビューから実装完了まで完全自動化（設計レビュー→作業計画→TDD実装）.
dashboard
Generar dashboard HTML local con métricas de eficiencia del proyecto SDD.
usage-add
PitWay: Accumulate measured planning or qa token usage onto a milestone.
hub-tdd
TDD workflow for MCP Hub implementation. Types → Tests (red) → Implementation (green) with git gates.
eval
Evaluate and improve one healthcare agent's system prompt. Run up to 5 iterations of: prepare fixed questions -> answer -> judge -> improve -> re-score -> commit if better.