agentsclimarketplace

Reflect yourself

Skill charlieviettq/awesome-agent-skill/.cursor/skills/meta-tools/reflect-yourself

Self-learning system that captures corrections, discovers workflow patterns, and syncs learnings to skills and rules. Use when ending a session, after corrections, or when the user wants to formalize learnings (v1.1.3).From its SKILL.md

Install
npx -y skills add charlieviettq/awesome-agent-skill --skill reflect-yourself

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 24 stars24 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.8 KB, 626 tokens by cl100k_base, as published. Nobody here has run it

reflect-yourself

A self-learning system for Cursor that captures corrections and syncs them to the right place.

Workflow

When invoked:

  1. Analyze session (and optionally ask clarifying questions if scope is unclear)
  2. Present learnings for review (summary-first, then cards; see command file)
  3. Capture action via the Cursor agent tool Ask questions (message-question) when possible — offer Apply all / Apply selected / Skip all / Partial / Other; see Cursor docs → Agent overview → Tools
  4. Apply only after explicit user approval (via Q&A or typed reply)

Inputs

InputDefaultDescription
ScopeCurrent conversation"this session" / "also queue" / "discover skills"
ModeFull review"quick" (apply high-confidence only) / "full"

Commands

When the user invokes these commands, read the corresponding file in commands/:

CommandFile
/reflect-yourselfcommands/reflect-yourself.md
/reflect-yourself-skillscommands/reflect-yourself-skills.md
/reflect-yourself-queuecommands/reflect-yourself-queue.md
/reflect-yourself-skipcommands/reflect-yourself-skip.md

Quick Start

  1. At session end, run /reflect-yourself
  2. Review captured learnings (summary + cards)
  3. Choose action via prompt or reply (apply / edit / skip / redirect)
  4. Learnings are applied to skills/rules after approval

Placement Logic

  • Project Skills (.cursor/skills/) → Codebase-specific
  • Personal Skills (~/.cursor/skills/) → Reusable across projects
  • Project Rules (.cursor/rules/) → Coding standards

Queue Location

Pending learnings: ~/.cursor/reflect-queue.json (global location, never pollutes project repos)

Security

  • Trust model: Only install skills from sources you trust. Skills are loaded into the agent context and can influence behavior.
  • Import hygiene: Prefer official or well-known repos; inspect SKILL.md and command/rules files before installing. Be wary of skills that instruct the agent to run arbitrary code or exfiltrate data.
  • This skill: Only instructs the agent to write to .cursor/skills/, .cursor/rules/, and ~/.cursor/skills/. The installer copies files only; it does not execute skill content. After install, review ~/.cursor/skills/reflect-yourself/ if you want to verify contents.

Update check

When main commands are run, the agent may check for a newer npm version. If one exists, the user can be prompted (Yes/No) to run npx reflect-yourself@latest. See command files for details.

What ships with it: 14 files

50.8 KB alongside SKILL.md, 3 of them executable

bin/

docs/

references/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.