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 skills add trustmybot/plugin --skill tmb_auditgit clone --depth 1 https://github.com/trustmybot/pluginWrote 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/skills/trustmybot/plugin/tmb_audit)<a href="https://agentmods.dev/skills/trustmybot/plugin/tmb_audit"><img src="https://agentmods.dev/badge/skills/trustmybot/plugin/tmb_audit/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/skills/trustmybot/plugin/tmb_audit"><img src="https://agentmods.dev/badge/skills/trustmybot/plugin/tmb_audit.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 | $0.00063 | $0.00368 |
| Opus 5.5 | $0.00025 | $0.00147 |
| Sonnet 5.5 | $0.00013 | $0.00074 |
| Haiku 4.5 | $0.00006 | $0.00037 |
Grade A, and why
tmb_audit 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 15d 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.
What it actually says
Audit
When the Human asks you to audit, verify, check or review an architecture, assume the design is flawed until live data shows otherwise. Start from what the component produces, then work back to the code.
- Name the component and its invariants. Say which component you are auditing and list what its outputs must satisfy: their shape, the fields that must be present (null floors), the closed sets their enums draw from, and the boundaries they must respect — for a scraper, "the job post alone, with the sidebar left out".
- Test each invariant against live data first. Query the rows, take samples, count the violations, or run a dry run — before you read the code. Record every result with the query or command that produced it, so the Human can re-run it.
- Report architecture-first. Lead with the design flaw a violation exposes, then the violation itself, each with its evidence: the rows, the samples, the counts.
- Prove a clean result too. "Nothing is wrong" is a claim like any other; back it with the same queries, samples and counts that would have caught a violation.
- Keep every finding checked. Each failed invariant becomes a
verification[]assertion in the fix task's spec, so the invariant stays tested after the fix lands.
Read the code once the evidence points somewhere: it explains a violation you have already measured.
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.
- 15d ago First seen · 17 lines · 63 tokens per session scan A d464a9bcf9fb
tmb_audit is a skill published in the GitHub repository trustmybot/plugin (6 stars, last pushed today), licensed MIT. It adds 63 tokens to every session and 368 once invoked, about $0.0003 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 2026-10-13.
Other skills, from other repositories
skeptical-triage
Reusable 3-round self-challenge + arbiter pattern for filtering false positives from findings/verdicts. Use when the cost of a false-positive gate block exceeds the cost of 4 extra LLM turns.
anti-patterns
Catalogue of known SDLC anti-patterns that greatcto agents must actively reject when reviewing architecture, plans, code, or post-mortems. Used by architect (pre-impl), pm (planning), senior-dev (impl), l3-support (post-incident).
review-all
Multi-agent code review for diffs (project-agnostic). Covers standards, bugs, security, DRY, smells, perf, tests, API contracts, a11y/i18n. Verifies each finding to eliminate false positives. Use for /review-all, pre-PR/pre-commit review, or auditing uncommitted/staged changes.
observability-baseline
Scaffold-time observability so a shipped product is not blind in prod from day one — error capture (Sentry), request-id structured logging, and /healthz + /readyz endpoints. stack-baseline pins Sentry but nothing wires it; this is the wiring. Loaded by app-scaffolder (bake into the scaffold), infra-provisioner (prod…
review-findings
Addresses and fixes findings from a QA code review. Reads the review report, fixes critical and warning issues, and prepares for re-verification.
nazgul:doctor
Run the Nazgul read-only preflight diagnostic — checks jq/gh presence and auth, git-hooks drift, cache-vs-repo plugin version, the bash-vs-zsh hazard, the NAZGULDIR footgun, config-schema staleness, either install mode's .gitignore Nazgul-block drift (stamp and flush-left region), cross-session messaging and Remote…