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.
git clone --depth 1 https://github.com/ANcpLua/ancplua-claude-pluginsWrote 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/ancplua/ancplua-claude-plugins/mmt)<a href="https://agentmods.dev/commands/ancplua/ancplua-claude-plugins/mmt"><img src="https://agentmods.dev/badge/commands/ancplua/ancplua-claude-plugins/mmt/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/ancplua/ancplua-claude-plugins/mmt"><img src="https://agentmods.dev/badge/commands/ancplua/ancplua-claude-plugins/mmt.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00031 | $0.00928 |
| Opus 5.5 | $0.00012 | $0.00371 |
| Sonnet 5.5 | $0.00006 | $0.00186 |
| Haiku 4.5 | $0.00003 | $0.00093 |
Grade A, and why
mmt 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 19d 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/mmt — Mutation-Minded Testing
Runs the full behavior-first test-quality pass on $ARGUMENTS (a file,
directory, glob, or — if omitted — the current git diff).
Coverage is a floor. Kill power is the signal. This command will not add tests whose only purpose is to raise a number.
Scope resolution
$ARGUMENTSempty → usegit diff --name-onlyas the scope.$ARGUMENTSis a file or glob → use it.$ARGUMENTSis a directory → recurse into it.
If the scope contains no source files, stop and say so. Do not run the workflow on an empty set.
Phase 1 — Architecture review
Spawn the architecture-reviewer agent with the scope. Collect its ranked
list of structural smells that force weak tests.
Gate: if the reviewer returns findings ranked as blocking (a structural issue that makes any mutation-resistant test impossible in the target), pause the workflow and surface those findings first. Do not proceed to write tests against a broken shape.
Phase 2 — Test judgement
Spawn the senior-tester-judge agent with the scope. Collect per-test
verdicts (STRONG / ACCEPTABLE / WEAK / KILL) and the suite-level
triage table.
Phase 3 — Test rewrite
Spawn the expressive-verifier-improver agent with the judge's WEAK and
KILL entries. Rewrites are applied to the files directly (agent has
Edit and Write). For each rewrite, the agent names the specific
mutation the new test catches.
If any test case cannot be rewritten without a structural change, the agent hands it back to Phase 1 (architecture-reviewer) for that specific seam.
Phase 4 — Branch closure
Spawn the branch-coverage-implementer agent with the scope. It:
- Enumerates semantic branches.
- Confirms each is covered by a
STRONGorACCEPTABLEtest from Phase 2 or 3. - Writes new high-signal tests for the gaps.
- Runs the project's coverage (and mutation-testing tool if configured).
- Reports remaining branches by bucket: architectural / dead-code / unreachable.
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.
- 19d ago First seen · 112 lines · 31 tokens per session scan A 73790df43530
mmt is a command published in the GitHub repository ANcpLua/ancplua-claude-plugins (6 stars, last pushed 2d ago), licensed MIT. It adds 31 tokens to every session and 928 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-01.
Other commands, from other repositories
merge-work-packets
Verify all tracks complete, invoke merging-worktrees, run QA gates, report final integration status.
merge-worktree-execute
Phase 2: Sequential Round Merging - merge worktrees in dependency order with testing after each round.
test-audit
Writes a versioned TESTAUDIT.md; with --fix, quarantines the rot in gated, revertible commits. TRIGGER WHEN: auditing a test suite, measuring test health, finding dead or flaky or redundant tests, or diagnosing failure causes and proposing temporary quarantine. DO NOT TRIGGER WHEN: one module's tests are consolidated…
check
Quick health check (git, lint, unit tests).
merge-worktree-verify
Phases 4-5: Final Verification and Cleanup — run full test suite, verify contracts, clean up worktrees.
pre-pr
Run complete validation workflow before creating PR.