audit · localixai — Claude Code skill: settings audit
audit is a Claude Code skill that checks project settings for options that waste tokens or have no effect. It also reviews project instructions and builds a token-usage report.
MIT ★ 171 updated 2026-11-17
/plugin marketplace add localixai/tokendoctor
/plugin install tokendoctor@tokendoctorWhat audit does
audit runs tokendoctor-audit --json and sorts findings by severity. It reads project memory and instruction files, flags conflicting or unsafe directions, cites file paths and line numbers, and suggests replacement wording. It does not edit files.
The skill then runs tokendoctor-dashboard --open to create a local HTML report from transcripts on the machine. It uses usage figures, not message text, and makes no network requests. To inspect what actually loaded, it can run tokendoctor-selfcheck after the user agrees; that command starts claude -p on Haiku for one request. The audit compares configuration with the docs; it does not measure token savings.
What it helps with
- find Claude Code settings that are ignored or waste tokens
- spot conflicts across project memory and instruction files
- break down token usage across local transcripts
- check which hooks and skill tier actually loaded
Example prompts for the agent
- Audit the Claude Code settings in this project.
- Find conflicting directions in the project memory and instruction files.
- Build a token-usage report and relate its figures to the findings.
How to install audit
1Plugin marketplace
/plugin marketplace add localixai/tokendoctor/plugin install tokendoctor@tokendoctorThis installs the whole tokendoctor plugin with all its skills
2Skills CLI (Vercel)
npx skills add localixai/tokendoctor --skill audit -a claude-code3Manually: put the files in place
~/.claude/skills/audit/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/audit/
1Skills CLI (Vercel)
npx skills add localixai/tokendoctor --skill audit -a codex2Manually: put the files in place
.agents/skills/audit/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add localixai/tokendoctor --skill audit -a cursor2Manually: put the files in place
.cursor/skills/audit/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install localixai/tokendoctor skills/audit2Manually: put the files in place
.github/skills/audit/SKILL.md1Skills CLI (Vercel)
npx skills add localixai/tokendoctor --skill audit -a gemini-cli2Platform CLI
gemini skills install https://github.com/localixai/tokendoctor --path skills/auditManually: put the files in place
.devin/skills/audit/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about audit
How do I install audit in Claude Code?
Add the marketplace with /plugin marketplace add localixai/tokendoctor, then install the plugin with /plugin install tokendoctor@tokendoctor.
What does audit do?
It checks Claude Code configuration against the documentation, reviews project instructions, and creates a local token-usage report. It does not modify project files.
What is needed to check which settings actually loaded?
Running tokendoctor-selfcheck requires consent. It makes one request by starting claude -p on Haiku. An existing log can instead be read with tokendoctor-selfcheck --log <file>.
Source text
SKILL.md · MIT
--- name: audit description: Audit this Claude Code setup for settings that waste tokens or silently do nothing. disable-model-invocation: true allowed-tools: Bash(tokendoctor-audit *), Bash(tokendoctor-dashboard *), Bash(tokendoctor-selfcheck --log *), WebFetch(domain:code.claude.com), Read, Glob argument-hint: [project-dir] --- Audit the Claude Code configuration for the current project (or `$ARGUMENTS` if given). 1. Run `tokendoctor-audit --json` (add `--project <dir>` when an argument was given). It is read-only and reports the Claude Code version, what it found, and a list of findings, each with a severity, an explanation and a documentation link. 2. Present the findings grouped by severity: errors first (settings that are wrong or ignored), then warnings, then info. For each, give one or two sentences in plain words and the doc link. Do not restate the whole detail text. 3. Be honest about limits: the audit checks configuration against documented behaviour. It does not measure savings, so state no percentages. Point to `/usage` (usage breakdown and recommendations) and `/context` (what is loaded now) for measuring. 4. Say which findings have an automatic fix (those with a `fixes` list) and offer `/tokendoctor fix <finding-id>` for them. Never edit files from this skill. 5. Review what Claude is told to do, not only how it is configured: read this project's memory (`~/.claude/projects/<project path with / as ->/memory/`, its `MEMORY.md` and the files it lists) and the instruction files (`~/.claude/CLAUDE.md`, `~/.claude/rules/`, the project's `CLAUDE.md` and `.claude/rules/`). Use your own judgment: point out anything there that makes Claude spend more than the work needs, contradicts the settings or another instruction, is stale, or is unsafe to keep. Quote `file:line` and say how to rewrite it. Where an instruction is plainly the person's own choice, show what it costs and leave the choice to them. 6. **Read the docs as they are today.** The audit checked every settings key and `env` variable against Claude Code's current settings reference and variable list (`facts.docs` says when they were read; null means offline with no copy, so those checks were skipped: say so). To explain a finding beyond its detail, or to answer a question about a setting, read its `doc` link: every page is served as Markdown at that URL (`…/docs/en/<page>.md`), and `https://code.claude.com/docs/llms.txt` indexes them all. Prefer this to memory or to a mirror such as context7: it is the text of the current release, while mirrors re-index on their own schedule. 7. **See what really loaded.** Findings from a debug log (`facts.debugLog` gives its time) come from a `claude --debug` run of the last day. To check now, offer `tokendoctor-selfcheck`: it starts `claude -p` once in this project on Haiku, with no saved session and no MCP servers (one small request, a few cents at most), and reports which copy of tokendoctor loaded and in which tier, whether the built-in guard was seated, mods hooking compaction ahead of it, skipped hooks, the window Claude Code really compacts at, and a skill listing over its budget. Run it only after the person says yes; `tokendoctor-selfcheck --log <file>` reads a log they already have, at no cost. 8. If `facts.version` is null, say the Claude Code version could not be read and that rules which depend on a minimum version were still applied. 9. Build the usage dashboard: run `tokendoctor-dashboard --open`. It reads every transcript on this machine (usage figures only, never message text), writes one HTML file with no network use, and opens it: spend by day, by kind of token, model, reasoning effort, main thread and each subagent type, context size, project, hour of the week, and the largest sessions, with filters. Give the path it prints, and tie one or two of its numbers to the findings above (for example the share of spend that was the conversation re-read from the cache, or above 200K tokens of context).
© rsaryev · MIT · localixai/tokendoctor
Similar skills
SKILL.md skills catalog →audit by other authors comes first.