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 dwarvesf/dwarves-kit --skill gauntlet-proof-auditgit clone --depth 1 https://github.com/dwarvesf/dwarves-kitWrote 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/dwarvesf/dwarves-kit/gauntlet-proof-audit)<a href="https://agentmods.dev/skills/dwarvesf/dwarves-kit/gauntlet-proof-audit"><img src="https://agentmods.dev/badge/skills/dwarvesf/dwarves-kit/gauntlet-proof-audit.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 5.1 | $0.00177 | $0.01981 |
| Opus 5 | $0.00088 | $0.00991 |
| Sonnet 5 | $0.00035 | $0.00396 |
| Haiku 4.5 | $0.00018 | $0.00198 |
Grade A, and why
gauntlet-proof-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 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gauntlet proof-audit
Overview
Audit every committed gauntlet run record against its own persisted evidence and report the
result. This is the gauntlet-proof-audit instance of docs/patterns/audit-loop.md: a skeptic
pass over docs/verification/gauntlet/*/ROUNDS.md (and AB-ROUNDS.md), the eval corpus
/kit:gauntlet produces and lib/gauntlet/stats.sh projects into numbers. Neither of those
checks that a record's CLAIMS match its own EVIDENCE: a "clean=true" that never reconciles with
checker-output.txt, a findings count with no matching finding, a scrubbed key that leaked
anyway, a run-dir that drifted from the contract's naming. This skill never re-runs a probe and
never rewrites a historical record; it reports discrepancies for the operator.
The four slots (per the audit-loop pattern)
| Slot | This instance |
|---|---|
| Item set | git ls-files 'docs/verification/gauntlet/*/ROUNDS.md' 'docs/verification/gauntlet/*/*-ROUNDS.md' 'docs/verification/gauntlet/*/AB-ROUNDS.md' (committed records only; a room copy's own ROUNDS is untracked and never enters the set) |
| Contract | every CLAIM in the record is backed by its committed EVIDENCE: markers well-formed; recorded verdict == committed checker-output.txt; findings count reconciles with the rows; scrub clean (no resolved credential VALUE in committed evidence; a bare op:// pointer is allowed, not a leak); run-dir grammar conforms; a quoted finding string is present in the cited transcript |
| Evidence class | Tier 1: the committed record files themselves via grep/wc/diff. Tier 2: kit:audit-scanner reads a quoted finding against the transcript. Evidence gitignored as a room artifact (transcript pruned) = UNTESTABLE → UNSURE |
| Apply mechanics | REPORT-first: a genuine discrepancy is FLAGGED with a dated correction note appended to the record (never a silent rewrite of the historical claim); UNSURE/DANGER listed in the PR body; REMOVE is disallowed (a record is evidence, not deleted) |
Process
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 · 100 lines · 177 tokens per session scan A 414625fb9ed2
gauntlet-proof-audit is a skill published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed today), licensed MIT. It adds 177 tokens to every session and 1,981 once invoked, about $0.0009 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 2026-09-04.
Other skills, from other repositories
verify-no-regressions
Verify no behavioral regressions after implementation. Spawns parallel subagents for semantic diff review + test execution. Use when: user says "verify no regressions", "check for regressions", "make sure nothing broke".
verify-against-rubric
Spawn a stateless sub-agent (Claude Code Task tool, Cursor agent mode, or equivalent) with ONLY the artifact path plus the locked rubric plus read-only tools. Sub-agent returns a structured verdict (satisfied, needsrevision, or failed) plus per-criterion feedback. Distinct from self-critique-and-revise (same-context…
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
verify-implementation
A workflow that runs a project’s verification skills to produce a report on coding patterns, architecture rules, and project conventions. It is intended for work after implementation, before a pull request, or during code review.
witness
Independently witness that an Allium loop's convergence claim is true and was reached honestly. Use when the user wants to verify a loop's self-report, confirm tests really pass and no generated test was weakened, produce a convergence certificate or witness record, gate CI on a trustworthy signal, or check that an…
ln-21-documentation-auditor
Audits documentation and comments for trustworthy claims, coverage, and discoverability. Not for code, test, or architecture audits.