policydb-activity-review

Scan audit_log entries and cluster unlogged user changes into sessions for review.

Updated Mar 23, 2026
One-click install
npx skills add https://github.com/wanderinglantern/policydb --skill policydb-activity-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: policydb-activity-review
Source: https://github.com/wanderinglantern/policydb/tree/main/.claude/skills/policydb-activity-review
Command: npx skills add https://github.com/wanderinglantern/policydb --skill policydb-activity-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Detects work sessions that occurred in audit trails but were not captured as logged activities, enabling accurate accountability and audit readiness. It clusters changes into sessions and writes unlogged user work to suggested_activities for review while excluding system-generated operations.

Core Features & Use Cases

  • Scan audit logs and cluster changes into sessions for review
  • Exclude system-generated operations to avoid false positives
  • Write unlogged user work to suggested_activities for review and follow-up
  • Support integration with anomaly_engineGate checks for recent activity and no-open-anomalies

Quick Start

Run the Activity Review Engine against the audit_log to generate suggested_activities for review.

Frequently Asked Questions about policydb-activity-review

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

FAQPage Schema
How do I detect unlogged user activity in audit logs for compliance reviews?

Audit log session clustering groups raw audit log changes into distinct user work sessions using configurable session gaps. This mechanism isolates specific user activity windows for accurate accountability and thorough audit reviews.

How do I exclude system-generated operations during an activity review?

Yes, you can configure session gaps and gating logic within the anomaly_engine. This allows you to customize how the audit log is clustered into sessions and how recent activity gates are checked during the activity review process.

What is the best way to prepare audit log data for an unlogged activity review?

The best way to prepare for an unlogged activity review is to ensure the audit_log contains complete records of user and system changes. The Skill processes these raw audit entries end-to-end to identify suspicious sessions and write suggested_activities.

Why does my recent activity gate fail when investigating suspicious sessions?

Your recent activity gate may fail because unlogged user activity was not captured in the standard logs. By scanning the audit_log and clustering changes, the Skill identifies missing sessions and writes them to suggested_activities to resolve gate check failures.