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 OutlineDriven/outline-driven-development --skill claude-settings-auditgit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote 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/outlinedriven/outline-driven-development/claude-settings-audit)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/claude-settings-audit"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/claude-settings-audit.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00044 | $0.01186 |
| Opus 5 | $0.00022 | $0.00593 |
| Sonnet 5 | $0.00009 | $0.00237 |
| Haiku 4.5 | $0.00004 | $0.00119 |
Grade A, and why
claude-settings-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 2d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- claude-settings-audit — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent command policy audit
Contract
| Field | Bound contract |
|---|---|
| Trigger | User sets up a project, audits agent command permissions, or asks which read-only bash commands and domains to allow |
| Authority | Read-only. No file, VCS, credential, paid, published, deployed, or remote mutation. |
| Side effect | Emits a recommended command and domain allowlist as chat output. Writes nothing to disk |
| Done | A validated, least-privilege command and domain policy recommendation containing only read-only, project-relevant commands and domains, with no state-modifying commands |
Inputs
- The repository root to audit (defaults to the current working directory).
- Optional: an existing policy file to merge into.
Procedure
- Parse repository manifests to identify the tech stack. List the repository root and find manifest files to depth 2 (
*.toml,*.json,*.lock,*.yaml,*.yml,Makefile,Dockerfile,*.tf). Classify by indicator files:- Python:
pyproject.toml,setup.py,requirements.txt,Pipfile,poetry.lock,uv.lock - Node.js:
package.json,package-lock.json,yarn.lock,pnpm-lock.yaml - Go:
go.mod,go.sum; Rust:Cargo.toml,Cargo.lock; Ruby:Gemfile,Gemfile.lock - Java:
pom.xml,build.gradle,build.gradle.kts - Build:
Makefile,Dockerfile,docker-compose.yml; Infra:*.tf,kubernetes/,helm/ - Monorepo:
lerna.json,nx.json,turbo.json,pnpm-workspace.yamlDone when: the tech stack is classified from detected manifest files.
- Python:
- Read any existing policy files. Tolerate absence. Done when: existing policy is read or confirmed absent.
- Synthesize a read-only command and domain allowlist specific to the detected stack. Build the baseline read-only commands, each as
Bash(<cmd>:*):ls,pwd,find,file,stat,wc,head,tail,cat,tree,git status,git log,git diff,git show,git branch,git remote,git tag,git stash list,git rev-parse,gh pr view,gh pr list,gh pr checks,gh pr diff,gh issue view,gh issue list,gh run view,gh run list,gh run logs,gh repo view. Add stack-specific read-only commands only for tools actually detected by lock files or manifests. Done when: the stack-specific allowlist is built. - Filter the allowlist to strictly forbid state-modifying commands. Remove any command that can mutate state: no install, build, run, write, delete, or push. Remove unrestricted API wrappers (e.g.
gh apiwithout a read-only subcommand) that can issue mutating requests. Include only the package manager the project actually uses: ifpnpm-lock.yamlis present, omit npm and yarn; ifyarn.lock, omit npm and pnpm; ifpackage-lock.json, omit yarn and pnpm. Where multiple lock files coexist, include commands for each detected manager. Done when: every remaining command is read-only, detected, and scoped. - Add
WebFetch(domain:...)entries for detected frameworks: Django todocs.djangoproject.com; Flask toflask.palletsprojects.com; FastAPI tofastapi.tiangolo.com; React toreact.dev; Next.js tonextjs.org; Vue tovuejs.org; Express toexpressjs.com; Rails toguides.rubyonrails.org,api.rubyonrails.org; Go topkg.go.dev; Rust todocs.rs,doc.rust-lang.org; Docker todocs.docker.com; Kubernetes tokubernetes.io; Terraform toregistry.terraform.io. Done when: framework domain entries are added for detected frameworks. - Format the recommendation as a safe policy block. Use the
:*suffix so a base command accepts any arguments. Never include absolute paths, user-specific paths, or project scripts that may have side effects. Done when: the policy block is formatted with only read-only, detected, scoped commands and domains.
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.
- 2d ago Changed · -30 tokens per session aa45a7d7c5b9
- 4d ago First seen · 48 lines · 74 tokens per session scan A 07cdec6c2f18
claude-settings-audit is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed 2d ago), licensed Apache-2.0. It adds 44 tokens to every session and 1,186 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
audit-project
Run an iterative multi-agent code audit until critical and high findings are resolved. Use when the user says "audit my code", "find all the bugs", "deep code audit", "iterative review", or "review until clean".
duet
Use when the user invokes /duet, says "pair with me", or faces aesthetic, architectural, or irreversible decisions.
goal-prompt-drafting
Use when asked to draft copy-ready /goal objectives for long-running agents; returns one normalized one-line objective with measurable end state, grounded proof, easy-out invariants, a stop clause, and a Missing list. Not for source or remote-system changes.
handoff-prompt
Use when the user asks for a handoff, delegation, or clipboard-ready prompt for another agent: a standalone path-free prompt copied to the clipboard, confirmed by title. Not for session-snapshot briefs — use handoff; never remote, credential, publish, deploy, or irreversible.
publish-branch
Use when asked to publish the checked-out branch: commit and push it on whatever branch it is, the default branch included. Not for creating branches, PRs, force pushes, or pushing any other branch; when the request excludes the default branch, use commit-push-current.
drill
Use when a concept needs practising rather than explaining: run a scaffolded exercise from worked example to independent problem, quiz the learner, run spaced recall over what they cleared, or probe for the gaps blocking what they want next. For explanation, use explain-concept; for an end-to-end build, use capstone.