capture-audit

Detect drift across session decisions, memories, and plan artifacts.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/Jennyfirrr/workspace-template --skill capture-audit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: capture-audit
Source: https://github.com/Jennyfirrr/workspace-template/tree/main/claude-skills/capture-audit
Command: npx skills add https://github.com/Jennyfirrr/workspace-template --skill capture-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Mechanical drift-check verifying that decisions, memories, skills, and plan-amendments made in the current session were propagated to every place they need to be. It catches the "I decided X but never updated Y" failure mode. It runs as a pre-commit gate, a pre-handoff gate, and standalone to enforce structural discipline.

Core Features & Use Cases

  • Splits into a mechanical half (the tool) and a judgment half (the agent) to prevent false positives.
  • Mechanical checks include memory-index sync, plan-frontmatter completeness checks, decision-log artifact existence, sentinel matching, and cross-reference integrity.
  • Serves as a pre-commit gate, pre-handoff gate, and standalone verifier to prevent drift and ensure auditable consistency.

Quick Start

Run the capture-audit tool to validate drift across memory, plans, and decisions before commit.

Frequently Asked Questions about capture-audit

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

FAQPage Schema
How do I detect drift between decisions, memories, and plans before committing?▼

You can detect drift by running a pre-commit audit that mechanically verifies memory-index sync, plan-frontmatter completeness, and decision-log existence to catch inconsistencies before release.

What is a pre-handoff gate for verifying decision-log and plan artifacts?▼

A pre-handoff gate is a verification step that checks cross-reference integrity and sentinel matching across decisions, memories, and plans to ensure auditable consistency before transferring work.

How do I prevent false positives when auditing memory-index and plan drift?▼

You prevent false positives by splitting the audit into a mechanical half that checks sync and frontmatter, and a judgment half where the agent validates the findings before reporting.

Can I run a standalone drift check on my session's decisions and memory artifacts?▼

Yes, you can run a standalone verifier to identify and report drift between the current session's decisions, memories, and plan artifacts, outputting a structured report with remediation notes.

What does a structured drift report include when plan frontmatter is incomplete?▼

A structured drift report includes specific findings on memory-index sync failures and plan-frontmatter incompleteness, and surfaces targeted remediation notes for each detected inconsistency.

When should I not use a mechanical drift-check for decision and memory updates?▼

You should avoid using only the mechanical half without the agent's judgment, as relying solely on automated checks without verifying cross-reference integrity may miss contextual inconsistencies.