Self-referential AI development loops - iterate until completion
npx claudepluginhub jes5199/chief-wiggumImplementation 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.
jes5199
A fork of the ralph-wiggum plugin for Claude Code, implementing self-referential AI development loops.
This fork exists for two reasons:
Includes fixes from PR #12642:
```! auto-execute syntaxAdditional fixes in v1.1.0:
--completion-promise --iterations) now rejected with helpful error messages instead of silently misinterpreting argumentssystemMessage not being visible to the AI)Also adds clearer documentation that completion promises must be wrapped in <promise> XML tags.
You can start a Wiggum loop just by asking Claude naturally:
"Wiggum this until the tests pass" "Keep trying to fix the build" "Run in a loop until it works"
The plugin includes a skill file that recognizes these phrases and automatically starts an appropriate loop with sensible defaults.
/plugin marketplace add jes5199/chief-wiggum
/plugin install chief-wiggum
Then restart Claude Code to load the plugin.
Note: The Claude Code plugin interface can be buggy. If the above commands don't work, ask Claude to directly edit your config files to install or uninstall the plugin.
The Ralph Wiggum technique is an iterative development methodology based on continuous AI loops, pioneered by Geoffrey Huntley.
Core concept: Feed Claude the same prompt repeatedly. Each iteration, Claude sees its previous work in files and git history, allowing it to iteratively improve until the task is complete.
This plugin implements Ralph using a Stop hook that intercepts Claude's exit attempts and feeds the same prompt back.
/wiggum-loop "Build a REST API for todos. Requirements: CRUD operations, input validation, tests. Output <promise>COMPLETE</promise> when done." --completion-promise "COMPLETE" --max-iterations 15
Or just ask naturally:
"Wiggum it - implement user authentication with tests"
Start a loop in your current session.
/wiggum-loop "<prompt>" --max-iterations <n> --completion-promise "<text>"
Options:
--completion-promise <text> - Phrase that signals completion (HIGHLY ENCOURAGED)--max-iterations <n> - Stop after N iterations (HIGHLY ENCOURAGED, recommend 10-20)IMPORTANT: Always specify BOTH options to prevent infinite loops.
Cancel the active loop.
/cancel-wiggum
To signal completion, output a <promise> tag with the exact text specified:
<promise>COMPLETE</promise>
The stop hook detects this tag and ends the loop. Always use both a completion promise AND --max-iterations (recommend 10-20) to prevent infinite loops.
Build a REST API for todos.
When complete:
- All CRUD endpoints working
- Input validation in place
- Tests passing (coverage > 80%)
- Output: <promise>COMPLETE</promise>
/wiggum-loop "Try to implement feature X" --completion-promise "FEATURE COMPLETE" --max-iterations 15
Always specify both --completion-promise and --max-iterations (recommend 10-20). This prevents infinite loops on impossible tasks.
Good for:
Not good for:
Official prompts.chat marketplace - AI prompts, skills, and tools for Claude Code
Open Design — local-first design app exposed to coding agents over MCP. Install once with your agent's plugin command and projects/files/skills are reachable through stdio.
Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations