Signed Audit Trails Recipe

This is a teaching cookbook for adding cryptographically signed audit trails to Claude Code tool calls. Every bash command, file edit, or web fetch gets evaluated against a Cedar policy, then signed with Ed25519 into a tamper-evident receipt. The receipts chain together and verify offline with a single CLI command, no network or vendor lookup needed. Most useful in regulated environments (finance, healthcare), CI/CD pipelines where you need proof a policy gate held, or when a counterparty wants to verify your agent's behavior without trusting you. The actual runtime lives in a separate protect-mcp plugin; this skill walks through the pattern, cryptography, and cross-implementation interop before you commit to the hooks.

skills.sh installs: 6.7K
GitHub stars: 42.4K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add wshobson/agents --skill signed-audit-trails-recipe --agent claude-code

Installs into .claude/skills of the current project.

Files

Signed Audit Trails for Claude Code Tool Calls

Cookbook-style walkthrough for cryptographically signed receipts on every Claude Code tool call. This is the teaching skill. For the runtime implementation, install the protect-mcp plugin.

What this gives you

Every tool call (Bash, Edit, Write, WebFetch) is:

Evaluated against a Cedar policy before execution. If the policy denies the call, the tool does not run.
Signed as an Ed25519 receipt after execution. Receipts are JCS-canonical and verifiable offline by anyone with the public key.

An auditor, regulator, or counterparty can verify every receipt later (Step 5). No network call, no vendor lookup, no trust in the operator.