One-click install
npx skills add https://github.com/hendrax5/ironman --skill audit-hendrax5
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: audit
Source: https://github.com/hendrax5/ironman/tree/main/skills/audit
Command: npx skills add https://github.com/hendrax5/ironman --skill audit-hendrax5

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provide a quick, consistent way to inspect audit logs, decision traces, and session history so teams can understand past AI-driven actions, verify phase transitions, and investigate anomalies for accountability and debugging.

Core Features & Use Cases

  • Recent Log Viewing: Read daily JSONL audit files and show the last 20 entries in reverse chronological order for immediate incident review.
  • Decision Trace Inspection: Filter and display chronological decision chains for a given feature to reveal rationale, alternatives, and confidence scores.
  • Search & Summary: Search by action, feature, or date and produce aggregated daily/weekly summaries and key metrics for compliance reporting.
  • Use Case: An engineer or auditor investigating an unexpected phase transition can run a trace for the affected feature, review the decision rationale, and cross-check related audit entries.

Quick Start

Ask the audit skill to display the latest 20 audit entries by invoking /audit log.

Frequently Asked Questions about audit

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

FAQPage Schema
How do I view recent audit logs in JSONL format for AI session history?

To view recent audit logs, the system reads daily JSONL audit files and displays the last 20 entries in reverse chronological order. This provides immediate visibility into recent AI actions and session history for incident review.

How can I trace decision rationale and confidence scores for a specific feature?

You can trace decision rationale by filtering chronological decision chains for a given feature. This inspection reveals the decision rationale, evaluated alternatives, and confidence scores to verify phase transitions and investigate anomalies.

Can I search audit logs by action, feature, or date for compliance reporting?

Yes, you can search audit logs by action, feature, or date. The system filters and aggregates entries to produce daily or weekly summaries and key metrics specifically formatted for compliance reporting.

What is the best way to investigate unexpected phase transitions in automation decisions?

The best way to investigate unexpected phase transitions is to run a decision trace for the affected feature. You can review the automation decision rationale and cross-check related audit entries to ensure accountability.

Do I need any specific dependencies to parse JSONL audit and decision trace files?

No specific dependencies are required to parse JSONL audit and decision trace files. The system independently reads, filters, and aggregates entries to return formatted recent logs, trace chains, or summarized metrics.