frankbria/ralph-claude-code

Autonomous AI development loop for Claude Code with intelligent exit detection

View on GitHub ↗Jump to charts ↓

Summary Information

Updated 56 minutes ago
Added to GitGenius on January 12th, 2026
Created on August 27th, 2025
Open Issues & Pull Requests: 33 (+0)
Number of forks: 721
Total Stargazers: 9,605 (+0)
Total Subscribers: 45 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 2.9 days
Mean response time: 26.6 days
90th percentile: 84.1 days
Tracked items: 154

How this project is maintained

Around half of the issues opened in the past year never receive a reply. 32% of open issues come from outside the core team, a mix of external reports and the maintainers' own roadmap. Work labelled "bug" is answered fastest, typically in under an hour, while "testing" waits about 3 months. Only 9% of issues opened in the past year have been closed. Three people close 97% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 28
New in 7 days: 0
Closed in 7 days: 0
Avg open age: 17 days
Stale 30+ days: 24
Stale 90+ days: 6

Recent activity

Opened in 7 days: 0
Closed in 7 days: 0
Comments in 7 days: 0
Events in 7 days: 0

Top labels

  • enhancement (80)
  • bug (39)
  • multi-provider (16)
  • priority: P1 (16)
  • priority: P2 (15)
  • priority: P4 (13)
  • priority: P3 (12)
  • testing (12)

Most active issues this week

No issue events were indexed in the last 7 days.

Detailed Description

Ralph is a shell-based autonomous AI development loop for Claude Code that enables continuous iterative improvement of projects with intelligent safeguards against infinite loops and API overuse.

Ralph addresses the problem of manual iteration in AI-assisted development by automating cycles where Claude Code progressively refines a project toward completion. It implements a technique that uses a dual-condition exit gate requiring both completion indicators and an explicit exit signal, preventing runaway loops. The tool includes rate limiting with hourly reset, a circuit breaker with advanced error detection, and a response analyzer with semantic understanding. It supports session continuity through a resume flag, configurable session expiration, and dry-run mode for testing without API calls. Additional features include live streaming output, log rotation, metrics tracking, desktop notifications, automatic git backup with rollback capability, and GitHub issue import with metadata filtering.

Developers should choose this tool if they want to automate repetitive Claude Code interactions while maintaining control over execution. It suits projects where iterative refinement is needed but manual intervention between cycles is undesirable. The tool is designed to be installed once globally and used across any directory, with project-specific configuration through a .ralphrc file and an interactive enablement wizard.

The project's issue tracker is primarily driven by internal planning rather than external user demand, with most issues raised by the maintainers themselves. Responses to new issues and pull requests typically arrive within a few days. Work in the tracker concentrates on enhancement, bug fixes, and priority-one items, indicating active maintenance focused on core functionality and stability.