Self Refine Reflection
Systematic self-reflection and iterative output refinement for AI agents. Based on Madaan et al. (2023) Self-Refine, Shinn et al. (2023) Reflexion, and Andrew Ng's Reflection pattern. Use when outputs need multi-round critique-and-improve cycles, when initial quality is insufficient, or when complex tasks benefit from self-correction.
Install
npx skills add Thomaszhou22/self-refine-skill --skill self-refine-reflectiongit clone --depth 1 https://github.com/Thomaszhou22/self-refine-skill.git /tmp/self-refine-skill
mkdir -p ~/.claude/skills
cp -r /tmp/self-refine-skill/. ~/.claude/skills/self-refine-reflectionPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Self-Refine Reflection Skill > An AI agent skill for systematic self-reflection and iterative output refinement. > Based on Madaan et al. (2023), Shinn et al. (2023), and Andrew Ng's Reflection design pattern. ## Platform Auto-Detection At skill load time, detect your runtime environment and adjust capabilities: | Capability | How to Check | Fallback |-----------|-------------|----------| | File system | Can you read `references/reflection-templates.md`? | Use the inline templates below instead | Persistent memory | Can you write to `memory/`? | Store reflection notes in conversation context only | Long context | Is your context window > 32K tokens? | Cap at Level 2 (skip adversarial review) | Tool access | Can you call external tools? | Use mental verification only **Detection rules:** - If you can read this file's `references/` directory → full mode (all levels + memory) - If you can read files but not write → full levels, in-conversation memory only - If you cannot read files at all → use inline templates (copied below), cap at Level 2 - If context is limited (< 8K usable) → default to Level 1, max Level 2 This means **every platform gets the best possible experience automatically** — no manual configuration needed. ## Inline Reflection Templates (for environments without file access) If you cannot read `references/reflection-templates.md`, use these directly: ### Level 1 Internal Prompt
Details
- Skill name
self-refine-reflection- Repository
- Thomaszhou22/self-refine-skill
- License
- MIT
- Repo updated
- 21 May 2026
FAQ
What does the Self Refine Reflection skill do?
Systematic self-reflection and iterative output refinement for AI agents. Based on Madaan et al. (2023) Self-Refine, Shinn et al. (2023) Reflexion, and Andrew Ng's Reflection pattern. Use when outputs need multi-round critique-and-improve cycles, when initial quality is insufficient, or when complex tasks benefit from self-correction.
How do I install the Self Refine Reflection skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add Thomaszhou22/self-refine-skill --skill self-refine-reflection. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Self Refine Reflection popular?
It lives in Thomaszhou22/self-refine-skill, which has 11 GitHub stars and 0 forks.
Related skills
Suno Songwriter
11regiellis/suno-songwriter-agent-skill
Write, revise, plan, or adapt songs, lyrics, hooks, and AI music prompts. Use for Suno vocal or instrumental packages, song matrices, multilingual adaptation, genre treatments, reusable artist personas, album plans, Suno settings and iteration, sound samples, and cover-art prompts. Supports original style-inspired writing without copying lyrics. Not for standalone music theory, general DAW mixing, or licensing advice.
Mcd Baby Bite
11JimWen/mcd-baby-bite
带娃吃麦当劳的宝宝餐定制官（小小麦门）。当用户提到带宝宝 / 孩子 / 幼儿（1-6 岁）去麦当劳、宝宝能吃什么、怎么去酱去盐、开心乐园餐怎么点、给店员怎么说、宝宝外卖备注、吃完附近哪里遛娃、生成宝宝麦当劳打卡照片、做中英文识物卡片、把带娃吃麦当劳分享到小红书时使用。基于麦当劳 MCP 门店实时菜单与官方营养数据，自动去掉线上可去的配料，按国标年龄档位算钠预算与份量，生成店员话术卡；可选高德（天气、步行到店、饭后遛娃）与小红书（笔记草稿 / 发布）。
Cli2skill
92233admin/cli2skill
Turn any CLI or MCP server into an Agent Skill. Use when you want to replace an MCP server with a zero-overhead CLI skill, or generate a skill from any command-line tool's --help output.
Idea Sparring
11harims95/straight-talk
Makes Claude a sparring partner instead of an agreement engine when evaluating ideas, plans, or decisions.
Codex Harness MCP
9chapzin/codex-harness-mcp
Use this skill when a user wants Codex CLI or another MCP-compatible coding client to work through a harness-engineering loop with explicit execution contracts, persistent local knowledge/RAG, durable traces, structured verification evidence, project-local governance policy, PASS/FLAG/BLOCK governance audits, trace-level observability reports, harness profiles, eval cases/runs, Meta-Harness-lite proposals and promotion decisions, natural-language harness spec export, MCP resources/prompts, compact handoff context, multi-client setup, and completion gates before claiming work is done. Triggers
Deglaze
11LuciferDono/deglaze
Strip the sycophancy. Apply accountability pressure to push past an LLM's default declare-done glaze. Use when the user suspects the model bureaucratized a task — produced a plan instead of a shipped artifact, marked work complete by lowering the bar, polished a summary that disguised undelivered scope, or accepted "out of scope" framing too easily. Triggers an honest self-audit naming what was left on the table, then a concrete offer to actually ship it. Trigger phrases (any tense, any phrasing) "did you do your best", "i bet you didn't", "i bet $X you", "what did you skip", "what did you lea
Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.