introspection-review-tracker

Persist adversarial code review findings as self_assessment memories via the llmem track-review CLI.

Updated Apr 27, 2026
One-click install
npx skills add https://github.com/MichielDean/LLMem --skill introspection-review-tracker
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: introspection-review-tracker
Source: https://github.com/MichielDean/LLMem/tree/main/skills/introspection-review-tracker
Command: npx skills add https://github.com/MichielDean/LLMem --skill introspection-review-tracker

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reference documentation for the automated ReviewOutcomeTracker hook that persists adversarial code review findings as self_assessment memories. The hook fires automatically after every review — no manual steps required. This skill documents the mapping from severity tiers to error categories and the content format.

Core Features & Use Cases

  • Tracks and persists review findings as self_assessment memories to enable auditability and learning.
  • Documents severity-to-category mappings and the required memory content format for introspection workflows.
  • Provides guidance for post-review verification and integration with the llmem CLI.

Quick Start

Run llmem track-review after each adversarial code review to persist findings.

Frequently Asked Questions about introspection-review-tracker

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

FAQPage Schema
How do I persist code review findings for auditability and learning?

You persist code review findings by saving them as self_assessment memories. This tracks review outcomes automatically after adversarial code reviews to enable continuous learning and auditability.

What is the best way to track adversarial code review outcomes?

The best way to track adversarial review outcomes is using an automated hook that maps severity tiers to error categories. This enforces consistent tracking and persists findings directly to your memory store.

How does severity-to-category mapping work for self-assessment memories?

Severity-to-category mapping translates review severity tiers into standardized error categories. This enforces a consistent content format for self_assessment memories stored during introspection workflows.

Do I need the llmem CLI to track review findings?

Yes, you need the llmem track-review CLI to persist findings. You also need a valid review findings JSON and access to the memory store to complete the self_assessment memory creation.

When do I need to persist review findings as memories?

You need to persist review findings as memories after every adversarial code review. This automated post-review verification step ensures findings are tracked for future introspection and auditability.

What format is required for review findings JSON when using track-review?

The review findings JSON must conform to the severity-to-category mapping content format. This structured format ensures self_assessment memories are created correctly in the memory store by the llmem CLI.