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 skills/franzos/claude-plugins/auditnpx skills add franzos/claude-plugins --skill auditgit clone --depth 1 https://github.com/franzos/claude-pluginsWhat 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 | $0.00114 | $0.01039 |
| Opus 5 | $0.00057 | $0.00519 |
| Sonnet 5 | $0.00023 | $0.00208 |
| Haiku 4.5 | $0.00011 | $0.00104 |
Grade A, and why
audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auditing a Forseti deployment
Produce a ranked findings list, not a lecture. Check the live config against the deployment's own tooling and the operator guide, rank by severity, and let the user choose what to fix. Read the config that is actually deployed; do not assume the example files.
1. Locate the config and the deployment shape
Find the active config.toml ($FORSETI_CONFIG_PATH), infra/kratos/kratos.yml, infra/hydra/hydra.yml, and how it is run (single binary vs multi-instance, SQLite vs Postgres, behind a proxy or not). Rationale for each item below is in docs/operator-guide.md.
2. Run the built-in linter
forseti config-check --strict lints the Ory config and is the fastest source of real findings. Fold its output into the punch-list.
3. Walk the checklist
For each item, record pass/fail and severity. Highest-severity items first:
- 2FA enforcement (critical). In
kratos.yml, bothsession.whoami.required_aalandselfservice.flows.settings.required_aalmust behighest_available.settings.required_aal: aal1opens the factor-removal bypass (an attacker with the password or recovery email strips the second factor). Forseti cannot enforce this from its side. - Secrets rotated and unique. Hydra
secrets.systemand pairwise salt; Kratossecrets.cookieandsecrets.cipher(exactly 32 chars); Forseti[security].cookie_secretset to a stable ≥32-byte value (not the ephemeral per-boot key). None committed to the repo. - No placeholder secrets in the live config. Confirm no OAuth
client_secret, audit token, or SSH/SAML key was carried over from an example or playground config; rotate anything that was. - Audit webhook token.
[audit].webhook_tokenset and matching the bearer in everyweb_hookinkratos.yml(Forseti refuses to boot if empty, but confirm it is not the dev placeholder). - Internal listener not public.
[internal].bindon loopback only. It carries the audit receiver and the POSIX resolver (your directory); the network ACL is load-bearing once it leaves loopback. - Forwarded-for trust.
[proxy].trust_forwarded_for = trueonly behind a proxy that strips inbound forwarded-for headers, else IP spoofing bypasses rate limits. - TLS and headers. Reverse proxy terminates TLS and passes
X-Forwarded-Proto: https+X-Forwarded-Host; CSPframe_ancestors/X-Frame-Optionsset for clickjacking defense. - Webhook signing key.
data/webhook-signing-key.pempresent,0600, valid Ed25519 PKCS#8, and backed up (rotation is stop-replace-restart). - Admin surface.
[admin].allowed_emailsis the intended operator set and non-empty; confirm the two-tier operator/org-owner model is understood. Admin requires AAL2 in code. - Database and backups. Postgres (not SQLite) for multi-instance, or corruption results. Kratos and Hydra Postgres backed up (losing Hydra's JWKS invalidates every issued id_token).
- DCR hardening.
[oauth].dcr_reserved_namesdenylist populated and DCR rate limits set. - Metrics.
/metricsgated by license and[metrics].scrape_token, and served only on the internal listener. - Retention scheduled.
forseti audit-pruneandforseti unverified-prunerun on a timer/cron; they do not run automatically. - Log hygiene. Kratos
log.leak_sensitive_values: falseoutside dev; secrets supplied via env, not TOML; email creds never overtls = "none".
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.
- yesterday First seen · 40 lines · 0 tokens per session scan A 24af55a89d85
audit is a skill published in the GitHub repository franzos/claude-plugins (1 stars, last pushed 20d ago), licensed MIT. It adds 114 tokens to every session and 1,039 once invoked, about $0.0006 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…