reflect-yourself

Capture Cursor session corrections and route them to skills or rules.

23|2|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/maorfsdev/reflect-yourself --skill reflect-yourself
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reflect-yourself
Source: https://github.com/maorfsdev/reflect-yourself/tree/main
Command: npx skills add https://github.com/maorfsdev/reflect-yourself --skill reflect-yourself

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Allows Cursor to grow smarter by capturing corrections and routing them to the right persistent destinations (project skills, personal skills, or project rules), so improvements persist across sessions.

Core Features & Use Cases

  • Learn from corrections: capture user corrections and store them as repeatable learnings.
  • Discover workflow patterns: identify tasks that recur and could be formalized as skills.
  • Smart routing: route learnings to .cursor/skills/, ~/.cursor/skills/, or .cursor/rules/ with guided approval.

Quick Start

Run /reflect-yourself at the end of a session to capture learnings and begin applying them to skills and rules.

Frequently Asked Questions about reflect-yourself

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

FAQPage Schema
How do I save Cursor corrections so the AI agent remembers them in future sessions?

To save Cursor corrections, run the reflect-yourself command at the end of a session to capture user corrections and translate them into persistent actionable learnings stored in project skills, personal skills, or project rules.

Can I turn recurring workflow patterns in Cursor into reusable project rules automatically?

Yes, you can identify recurring tasks and formalize them as skills by running the reflection workflow, which discovers workflow patterns and routes them to .cursor/rules/ or .cursor/skills/ with guided user approval.

How does agent self-learning from session corrections actually work?

Agent self-learning works by capturing corrections from Cursor sessions and routing them to the right persistent destinations, allowing Cursor to grow smarter so that improvements persist across sessions without manual repetition.

Do I need a valid SKILL.md file to apply session learnings to my project skills?

Yes, applying session learnings requires a valid SKILL.md frontmatter with a name and a description, and it can include optional scripts, references, or assets for on-demand utilities to function properly.

What is the best way to route AI corrections to personal skills versus project rules?

The best way to route corrections is using the smart routing feature, which guides user approval to direct learnings to .cursor/skills/, ~/.cursor/skills/, or .cursor/rules/ based on whether they apply globally or to a specific project.

Will the agent modify my project rules without asking for approval first?

No, the agent requires explicit user approval before modifying skills or rules, ensuring you maintain control over which corrections are added to your persistent project or personal skill destinations.