memory-audit

Enumerate Claudia's knowledge graph entities, memories, commitments, documents, and relationships with provenance traces.

283|44|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/kbanc85/claudia --skill memory-audit-kbanc85
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: memory-audit
Source: https://github.com/kbanc85/claudia/tree/main/template-v2/.claude/skills/memory-audit
Command: npx skills add https://github.com/kbanc85/claudia --skill memory-audit-kbanc85

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Claudia's knowledge graph grows over time and its provenance is often unclear; this Skill provides an auditable view of everything Claudia knows, including entities, memories, commitments, documents, and relationships with source traces.

Core Features & Use Cases

  • All-entity audit: produce counts for entities, memories, commitments, documents, and relationships.
  • Per-entity profiling: drill into a single entity to surface memories, relationships, and linked documents.
  • Provenance tracing: show the source for each memory and claim to support verification.

Quick Start

Run /memory-audit to generate a full system audit.

Frequently Asked Questions about memory-audit

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

FAQPage Schema
How do I audit a knowledge graph to trace entity memory provenance?

A knowledge graph audit summarizes known entities, memories, commitments, documents, and relationships while providing provenance traces to verify the source of each claim and memory.

What is the best way to profile a single entity's memories and linked documents?

Per-entity profiling queries a single entity to surface its associated memories, relationships, and linked documents, presenting the results in structured tables with source traces.

Can I get a full system count of all memories, commitments, and documents?

Yes, an all-entity audit produces structured counts for every entity, memory, commitment, document, and relationship in the knowledge graph by querying memory APIs.

How does provenance tracing work for knowledge graph memories?

Provenance tracing shows the source for each memory and claim by querying memory APIs, returning source traces that support verification of the knowledge graph's contents.

What happens to the audit if document sources are missing?

The audit gracefully handles missing sources by continuing to gather available counts, profiles, and provenance data from the memory APIs without failing the entire query.

Do I need any dependencies to run a knowledge graph memory audit?

No dependencies are required; the audit relies on built-in memory APIs to gather counts, profiles, and provenance data, gracefully handling missing sources without external tools.