lsnackerman

lsnackerman / raw-clauding-skills

Public

workflows for compound thinking with AI

maven.comp94b9207-skills-for-compound-thinking-with-claude-code claude-skills
18
4
100% credibility
Found Feb 24, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A set of reusable markdown prompts designed as slash-command skills to structure AI conversations for better thinking workflows, session management, and task tracking.

How It Works

1
📰 Discover the skills

You hear about Raw Clauding Skills, a set of simple files that turn your AI chats into powerful thinking sessions.

2
📥 Grab the files

Download the collection of skill files to your computer.

3
📁 Set up your workspace

Create a few folders in your project for saving sessions and tasks as you go.

4
🔗 Add skills to your AI

Copy or place the skill files where your AI can see them, ready to use by name.

5
💬 Start a thinking session

In your AI chat, type a skill name like /session to capture your full conversation and ideas.

6
📝 Build and wrap up

Use skills like /task or /wrap to organize thoughts, make decisions, and end sessions neatly.

7
🔄 Pick up anytime

Next time, use /pickup to reload context and continue right where you left off.

🎉 Think like a pro

Your AI now partners with you through complete thinking cycles, sharper and more reliable every time.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is raw-clauding-skills?

Raw-clauding-skills provides markdown-based workflows for compound thinking with AI, packaged as slash commands like /session, /wrap, and /pickup that you symlink into Claude for repeatable sessions. A Python CLI extracts verbatim conversation logs from Claude Code jsonl files, dumping them into markdown for persistence in sessions/ or tasks/ folders. It tackles AI chat amnesia, creating a full lifecycle for context-heavy work without compression.

Why is it gaining traction?

These raw skills stand out by treating workflows like github workflows folder setups—drop-in, customizable paths mirroring github workflows environment variables for sessions and tasks. The hook is compounding power: /explore builds knowledge reserves over time, /gut pressure-tests ideas, outperforming one-off prompts in sustained AI partnerships. Cross-compatible with Cursor, ChatGPT, or any AI, they echo workflows github actions structure but for thinking.

Who should use this?

AI-reliant indie devs or researchers chaining Claude sessions for planning, note synthesis, or idea validation in Python projects. Perfect for solo workflows github projects where /pickup restores full context, or teams sharing task docs via BACKLOG.md. Skip if you're not deep into clauding daily.

Verdict

With 18 stars and 1.0% credibility, it's raw and immature—great README but no tests or broad adoption—yet forkable for custom skills. Try if compound AI thinking fits your github workflows runs on needs; otherwise, watch for polish.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.