Ralph Wiggum Plugin

Plugin Active

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.

Purpose

To enable continuous, iterative development by creating self-referential AI loops that allow Claude Code to autonomously refine its work until a task is completed.

Features

Continuous AI development loops
Self-referential prompt feedback
Iterative improvement based on file history
Configurable completion promises
Maximum iteration limits for safety

Use Cases

Developing complex features by breaking them down into iterative steps
Refactoring code and ensuring tests pass across multiple iterations
Automating the process of improving code quality and documentation
Tackling well-defined tasks where success criteria can be clearly articulated

Non-Goals

Replacing targeted debugging for immediate production issues
Performing one-shot operations that do not benefit from iteration
Handling tasks with inherently unclear success criteria
Making human judgment or design decisions

License

warning:License usabilityThe license is only mentioned in the README footer, with no dedicated LICENSE file or SPDX identifier in the manifest, making its usability unclear for downstream tools.

Trust

warning:Issues AttentionOpen issues (9462) far exceed closed issues (22831) in the last 90 days, indicating slow response times from maintainers.

Code Execution

info:LoggingThe stop hook script logs informative messages to stderr about loop status and potential issues, but does not write to a persistent audit log.