By ABConvert
Interactive self-referential development loop - runs the same prompt iteratively until completion
Official Claude Code plugin marketplace by ABConvert.
Using CLI:
claude plugin marketplace add ABConvert/abconvert-plugins-official
Using Claude Code directly:
/plugin marketplace add ABConvert/abconvert-plugins-official
Plugins can be installed in different scopes:
| Scope | Description |
|---|---|
user | Available in all your Claude Code sessions (default) |
project | Only available in the current project |
Using CLI:
# Install for user (default)
claude plugin install ralph-wiggum@abconvert-plugins
# Install for current project only
claude plugin install ralph-wiggum@abconvert-plugins --scope project
Using Claude Code directly:
# Install for user (default)
/plugin install ralph-wiggum@abconvert-plugins
# Install for current project only
/plugin install ralph-wiggum@abconvert-plugins --scope project
An iterative development loop that runs Claude with the same prompt repeatedly until task completion. Based on the Ralph Wiggum technique pioneered by Geoffrey Huntley.
How it works:
Best for:
Not recommended for:
| Command | Description |
|---|---|
/ralph-wiggum:help | Show help documentation |
/ralph-wiggum:ralph-loop | Start a Ralph loop |
/ralph-wiggum:cancel-ralph | Cancel an active loop |
# Basic loop with iteration limit
/ralph-wiggum:ralph-loop "Build a REST API for todos" --max-iterations 20
# Loop with completion promise
/ralph-wiggum:ralph-loop "Fix all TypeScript errors" --completion-promise "All tests passing"
# Combined options
/ralph-wiggum:ralph-loop "Refactor auth module" --max-iterations 50 --completion-promise "DONE"
# Cancel an active loop
/ralph-wiggum:cancel-ralph
For fully autonomous loops where you don't want to approve each action, start Claude Code with the --dangerously-skip-permissions flag:
# Start Claude Code in autonomous mode
claude --dangerously-skip-permissions
# Then run your ralph loop
/ralph-wiggum:ralph-loop "Build a REST API" --completion-promise "DONE" --max-iterations 50
Use cases for autonomous mode:
Warning: This skips all permission prompts. Only use in trusted environments with well-defined tasks and iteration limits.
To exit the loop, Claude must output the exact promise text in XML tags:
<promise>YOUR_PROMISE_TEXT</promise>
The promise must be genuinely true - Claude cannot lie to escape the loop.
Using CLI:
# Remove a plugin
claude plugin uninstall ralph-wiggum@abconvert-plugins
# Remove the marketplace
claude plugin marketplace remove abconvert-plugins
Using Claude Code directly:
# Remove a plugin
/plugin uninstall ralph-wiggum@abconvert-plugins
# Remove the marketplace
/plugin marketplace remove abconvert-plugins
This version includes fixes from PR #12642 addressing issue #12170:
Issues and pull requests welcome at github.com/ABConvert/abconvert-plugins-official.
MIT
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub abconvert/abconvert-plugins-official連續自引用 AI 循環,用於互動式迭代開發。實現 Ralph Wiggum 技術,讓 Claude 在 while-true 循環中執行相同提示直到任務完成
Implementation of the Ralph Wiggum technique - continuous self-referential AI loops for interactive iterative development. Run Claude in a while-true loop with the same prompt until task completion.
Interactive self-referential AI loops for iterative development. Claude works on the same task repeatedly, seeing its previous work, until completion.
Self-referential development loop plugin - iteratively improves on tasks until completion
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.