reflexion:memorize

Curate reflections and critique outputs into structured CLAUDE.md entries.

2|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/fockus/claude-skill-build --skill reflexion-memorize-fockus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reflexion:memorize
Source: https://github.com/fockus/claude-skill-build/tree/main/skills/reflexion-memorize
Command: npx skills add https://github.com/fockus/claude-skill-build --skill reflexion-memorize-fockus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Curates insights from reflections and critiques into CLAUDE.md to build a living playbook using Agentic Context Engineering (ACE).

Core Features & Use Cases

  • Automated curation: transforms reflections and critique outputs into structured CLAUDE.md entries for easy retrieval and reuse.
  • Context growth and refinement: applies ACE growth-and-refine rules to evolve knowledge over time without context loss.
  • Quality and safety controls: enforces non-redundancy, specificity, and verifiability for each memorized insight.

Quick Start

Memorize the latest reflections and critique outputs and update CLAUDE.md to grow a living ACE playbook.

Frequently Asked Questions about reflexion:memorize

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

FAQPage Schema
How do I save reflections and critique outputs into CLAUDE.md?

To save reflections into CLAUDE.md, you can curate critique outputs into structured entries using ACE growth-and-refine rules. This process organizes insights into actionable bullets that evolve your knowledge base without context loss.

What is Agentic Context Engineering for memory management?

Agentic Context Engineering (ACE) is a method to grow a living playbook by transforming reflections into structured CLAUDE.md entries. It applies growth-and-refine rules to evolve knowledge over time while ensuring memos are verifiable and non-redundant.

How do I prevent redundant entries when updating a knowledge playbook?

To prevent redundant entries when updating a knowledge playbook, apply ACE quality controls that enforce non-redundancy, specificity, and verifiability for each memorized insight. This ensures all bullets remain actionable and clearly sourced.

What's the best way to structure insights in CLAUDE.md?

The best way to structure insights in CLAUDE.md is to apply ACE rules that produce structured, actionable bullets organized by specific sections. This includes attaching clear sources and confidence levels to each memo.

Does memorizing reflections into CLAUDE.md require any dependencies?

Memorizing reflections into CLAUDE.md using ACE rules requires no external dependencies. The process operates independently to curate insights, apply safety controls, and generate structured playbook entries.