whatisgithub

What is woodshed?

gaearon/woodshed — explained in plain English

Analysis updated 2026-07-17 · repo last pushed 2026-01-13

★ 65TypeScriptAudience · pm founderComplexity · 3/5QuietSetup · moderate

In one sentence

An experimental tool for building and testing AI skills for Claude: write instructions and test cases, run them, have Claude grade the results, then iterate.

Mindmap

mindmap
  root((woodshed))
    What it does
      Create Claude skills
      Run test prompts
      Auto grade results
    Tech stack
      TypeScript
      Claude API
    Use cases
      Prototype AI skills
      Test customer service replies
      Iterate on instructions
    Audience
      Founders
      Non technical experimenters
    Caveats
      Early alpha software
      Uses Claude API tokens
      Rough edges expected

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Create a skill as plain-text instructions and test it against multiple real-world prompts.

USE CASE 2

Automatically grade how well Claude followed a skill's instructions using another Claude instance.

USE CASE 3

Iterate on a customer service reply skill by reviewing saved results and refining instructions.

USE CASE 4

Review saved logs of prompts, responses, and grades to learn what worked and what didn't.

What is it built with?

TypeScriptClaude API

How does it compare?

gaearon/woodsheddenissergeevitch/chatgpt-portaljustjake/sqlite3-parser-js
Stars656664
LanguageTypeScriptTypeScriptTypeScript
Last pushed2026-01-13——
MaintenanceQuiet——
Setup difficultymoderatemoderateeasy
Complexity3/53/52/5
Audiencepm founderdeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Experimental alpha-stage tool that can consume significant Claude API tokens.

Copy-paste prompts

Prompt 1
Help me create a new woodshed workspace and write a skill for summarizing text in a specific style.
Prompt 2
Show me how to write test cases and evaluation criteria for a woodshed skill.
Prompt 3
Explain how woodshed uses a second Claude instance to grade the results of a skill run.
Prompt 4
What does woodshed save in its results folder after running a skill, and how do I review it?

Frequently asked questions

What is woodshed?

An experimental tool for building and testing AI skills for Claude: write instructions and test cases, run them, have Claude grade the results, then iterate.

What language is woodshed written in?

Mainly TypeScript. The stack also includes TypeScript, Claude API.

Is woodshed actively maintained?

Quiet — no commits in 6-12 months (last push 2026-01-13).

How hard is woodshed to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is woodshed for?

Mainly pm founder.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.

What are similar repos?