audit

Parse audit JSONL files into readable tables of permission decisions.

33|5|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/empathic/clash --skill audit-empathic
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit
Source: https://github.com/empathic/clash/tree/main/clash-plugin/skills/audit
Command: npx skills add https://github.com/empathic/clash --skill audit-empathic

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams quickly understand who granted or denied tool actions by viewing recent permission decisions from the audit log, enabling faster debugging and policy tuning.

Core Features & Use Cases

  • Audit review: Read the most recent entries from the audit log to assess recent activity.
  • Summarization: Convert raw JSONL entries into a readable table with time, decision, tool, input, reason, and matched rules.
  • Troubleshooting: Filter by tool name, decision type, or time range to diagnose policy behavior and edge cases.

Quick Start

Tail the last 20 audit entries from ~/.clash/audit.jsonl and display a readable summary.

Frequently Asked Questions about audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I view recent clash permission decisions from the audit log?

You can troubleshoot permission policy behavior by filtering the audit log entries by time, tool, or decision. This Skill parses raw JSONL entries to present a readable table for diagnosing policy edge cases and outcomes.

How do I read and summarize raw JSONL audit entries into a readable format?

This Skill summarizes raw JSONL audit entries by parsing the log data and converting it into a readable table that includes the timestamp, decision, tool, input, reason, and matched rules for quick review.

Can I filter clash audit logs by tool name or decision type?

Yes, you can filter clash audit logs by tool name, decision type, or time range to diagnose policy behavior. This allows you to isolate specific tool invocations and understand why permission decisions were granted or denied.

What is the best way to diagnose permission policy clashes and edge cases?

Diagnosing permission policy clashes involves reviewing the audit log to understand tool usage and outcomes. This Skill filters recent JSONL entries by time, tool, or decision to help identify policy edge cases.

Does this audit log reader require any specific dependencies to parse JSONL files?

No specific dependencies are required to parse JSONL audit files. This Skill independently reads the audit log, extracts timestamps and fields, and formats the permission decisions into a readable table for review.