Borrowing it
Nothing to install: this file belongs to starSumi/sumi-docs. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/starSumi/sumi-docs/main/.agents/skills/sumi-docs-audit/SKILL.mdgit clone --depth 1 https://github.com/starSumi/sumi-docsWrote 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/starsumi/sumi-docs/sumi-docs-audit)<a href="https://agentmods.dev/skills/starsumi/sumi-docs/sumi-docs-audit"><img src="https://agentmods.dev/badge/skills/starsumi/sumi-docs/sumi-docs-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/starsumi/sumi-docs/sumi-docs-audit"><img src="https://agentmods.dev/badge/skills/starsumi/sumi-docs/sumi-docs-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 | $0.00090 | $0.01749 |
| Opus 5 | $0.00045 | $0.00874 |
| Sonnet 5 | $0.00018 | $0.00350 |
| Haiku 4.5 | $0.00009 | $0.00175 |
Grade A, and why
sumi-docs-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 10d 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.
This is a copy
100% identical to sumi-docs-audit — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audit Sumi Docs
Return one compact, findings-first audit of the live target. Treat the audit as
read-only: do not modify source, configuration, Git state, GitHub state, or host
state; do not install or update tools. Repository validation may refresh only
its normal ignored build outputs. Return the report to the caller by default.
Write a report only when the user names an ignored artifacts/ destination.
Bind the evidence snapshot
Read the root and nearest nested AGENTS.md before selecting checks. Resolve
the target rather than inferring it from an earlier report:
git rev-parse --show-toplevel
git rev-parse HEAD
git branch --show-current
git status --porcelain=v2 --branch
node --version
pnpm --version
Record the UTC timestamp, operating system, target path, HEAD, branch, tree
cleanliness, tool versions, commands, exit codes, and skipped checks. Inspect
staged, unstaged, and untracked paths when the tree is dirty. Attribute results
to HEAD only for a clean tree; otherwise identify the audited snapshot as
HEAD plus the listed worktree changes and keep commit evidence separate from
worktree evidence.
Use current source, tests, generated artifact contents, GitHub checks, and live
probes in that order of relevance. Treat docs/history/ and package history
snapshots as non-authoritative. Redact secret values and machine-specific
absolute paths from the report.
Select the audit profile
- For a repository health audit, cover every applicable ownership boundary and run the root validation gates.
- For a release-candidate audit, also bind the candidate commit and artifacts, inspect the current candidate workflow and public checks, and verify the packaging, compliance, performance, publication, and rollback evidence that the candidate claims.
- For an architecture consistency audit, trace each affected contract from its authority through generated projections, consumers, tests, and operational documentation. Do not replace runtime evidence with an architecture diagram.
- For a PR, commit, or branch diff, use
$sumi-docs-pr. For a security-only scan, use the available security audit workflow. For a single failure, run a focused diagnosis rather than expanding it into a repository audit.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 161 lines · 90 tokens per session scan A d1c9bb84f5c4
sumi-docs-audit is a skill published in the GitHub repository starSumi/sumi-docs (1 stars, last pushed yesterday), licensed MIT. It adds 90 tokens to every session and 1,749 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to sumi-docs-audit, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
writing-emdash-docs
Write, revise, and review EmDash documentation for technical accuracy, useful structure, house style, and natural prose. Use when working on user documentation, guides, API reference pages, migration or upgrade guides, READMEs, contributor documentation, technical specifications, or release notes in the EmDash…
triage-contributor-pr
Triage open pull requests from external contributors to prisma/prisma and produce a per-PR verdict with evidence. Use when a maintainer asks to triage, evaluate, assess, or review the queue of incoming contributor PRs, to decide whether a fork PR is safe to run CI on, to check whether a PR is in scope for its version…
review-implement-phase
Implements triaged review actions, commits focused fixes, and posts Done plus resolves threads. Use when the user wants only the implementation phase of the review-framework workflow.
review-triage-phase
Produces canonical review actions from fetched review state and renders action markdown. Use when the user wants only triage/action-planning for the review-framework workflow.
update-pr
Update the pull request for the current session. Use when the user wants to push new changes to an existing PR.
review
Review one pull request for real bugs, regressions, and convention violations. Enumerate candidate issues across the whole diff, verify each against the code, then return structured line-anchored findings and a verdict. Read-only on GitHub; the orchestrator posts the review.