Auditrail — Security skill for Claude Code
The audit trail your coding agents already wrote, priced at list, ranked by what to fix, and scanned for leaked secrets, entirely on your own machine.
How to install Auditrail
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open 0xelitesystem/auditrail and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Auditrail does
The audit trail your coding agents already wrote, priced at list, ranked by what to fix, and scanned for leaked secrets, entirely on your own machine. It reads the session logs that are already there and tells you what the work was worth, where th...
Alternatives in Security
- Security Scan Report — Generated: 2026-04-10 20:48 UTC Skills scanned: 134 Total findings: 836 Critical: 32 High: 50 Safe skills: 100 18.1k ★
- ThinkWatch — Self-hosted AI API and MCP gateway for organizations: SSO and RBAC, per-user identity for MCP tool calls, PII 817 ★
- OpenTag — Open-source, channel-native agent gateway for Slack 499 ★
README
Auditrail
The audit trail your coding agents already wrote, priced at list, ranked by what to fix, and scanned for leaked secrets, entirely on your own machine. It reads the session logs that are already there and tells you what the work was worth, where the money went, and what to change. No account, no API key, no network call.

*Synthetic demo data, which is why the image itself says so. This card was generated from a seeded fake persona that ships with the repo (`test/fixtures/showcase/power-user`), not from anyone's real sessions. The marker is drawn only for this project's own showcase images (`card --demo-mark`); your card never carries it. Yours is built from your own logs and never leaves your machine.*
**[Open the live demo](https://0xelitesystem.github.io/auditrail/)** to see the full report with nothing installed. It is one HTML file with the same synthetic data inside it.
Install
npx auditrail
It reads `~/.claude/projects`, writes one self-contained HTML file, and opens it in your browser. It takes a few seconds, spends zero tokens, and calls no model. Nothing is uploaded, and there is nothing to sign up for.
`pnpm dlx auditrail` and `bunx auditrail` work too. Node 22 or later. Zero runtime dependencies.
Use
Run it with no arguments. It scans `~/.claude/projects`, writes one self-contained HTML file to `~/.auditrail/reports/`, and opens it in your browser. That is the whole loop, and for most people it is the only step. Read the fix list, change one thing, run it again next we
Related Skills
Agentconfig Doctor
MCP security scanner & config linter for AI coding agents. Finds leaked secrets, risky commands, broad permiss
Minga
Lean, security-first agent marketplace for Claude Code — a 3k-token core team (plan, build, verify, ship) plus
Artel
Lean, security-first agent marketplace for Claude Code — a 3k-token core team (plan, build, verify, ship) plus
Engineering Audit
Full-spectrum engineering health audit: scores Maintainability, Security, Testing, Architecture, Documentation
Diagnose Conversion Gap
Funnel teardown — stage-by-stage CVR vs benchmarks, $ leak waterfall, 40/40/20 root-cause (Audience/Offer/Copy
Compliance Audit Skill
A Claude skill that audits your codebase for GDPR and EU AI Act compliance and returns a ranked list of fixes.
Related Agents
Wh Eval Designer
Design-phase agent, tier 2 and above only. Reviews and extends the evals.md the designer wrote: adds the edge,
A0dev
Most people who want an app cannot navigate Xcode, signing certificates, or store submission, so a0.dev remove
Secret Scanner
PYA specialist spawned by the /pya orchestrator. Finds hardcoded and leaked credentials in an untrusted reposi