reflect

Distill session learnings into AGENTS.md, hooks, and memory artifacts.

13|3|Updated Mar 2, 2026
One-click install
npx skills add https://github.com/phrazzld/spellbook --skill reflect-phrazzld
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reflect
Source: https://github.com/phrazzld/spellbook/tree/main/skills/reflect
Command: npx skills add https://github.com/phrazzld/spellbook --skill reflect-phrazzld

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Session retrospectives, codification, and implementation feedback capture. Distill learnings to AGENTS.md/hooks/rules, propose lean improvements, tune repos for agents.

Core Features & Use Cases

  • Codifies recurring learnings into reusable artifacts (AGENTS.md, hooks, and policies).
  • Enables structured retro-processing, artifact generation, and repo tuning.
  • Use Case: turn a retrospection into concrete changes to agent behavior and project docs.

Quick Start

Run /reflect after a session to capture findings and generate codified artifacts.

Frequently Asked Questions about reflect

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

FAQPage Schema
How do I capture session learnings and update agent instructions automatically?

You can capture session learnings by running a retrospective to distill findings from git history and conversations, categorize them, and automatically update agent instructions in AGENTS.md.

What is the best way to codify recurring workflow learnings into reusable artifacts?

Codifying recurring learnings involves distilling session findings into structured artifacts, generating updated hooks, and applying repository tuning rules to ensure lean improvements across multiple projects.

How do I turn a development retrospection into concrete changes to project docs?

Turning a retrospection into concrete changes requires gathering evidence from your session, categorizing the findings, and producing codified artifacts like AGENTS.md updates and policies to tune agent behavior.

Does this retrospective workflow support gathering evidence from git history and conversations?

Yes, the retrospective workflow supports gathering evidence directly from git history and conversations to categorize findings and produce codified artifacts for your repositories.

Can I use session retrospectives to tune multiple project repositories for agents?

Yes, you can use session retrospectives to tune multiple project repositories by applying defined retro formats and codification rules to generate consistent agent instructions and artifacts.

When do I need to generate codified artifacts and hooks from session findings?

You need to generate codified artifacts and hooks when you want to turn recurring session learnings into automated repository policies and lean improvements for AI workflows.