Ralphy
A ralphy-implemented ralphy. This project was built using the original ralphy autonomous coding loop.
Autonomous AI Coding Loop that orchestrates AI coding assistants to implement tasks until completion.
Installation
pip install ralphy
Quick Start
# Initialize project config
ralphy --init
# Run a single task
ralphy "add user authentication"
# Run PRD until complete
ralphy --prd PRD.md
Features
- Multiple AI Engines: Claude Code, OpenCode, Cursor, Codex, Qwen-Code, Factory Droid
- PRD Mode: Loop through tasks from Markdown, YAML, or GitHub Issues
- Single Task Mode: Execute one task in an existing project
- Parallel Execution: Run tasks concurrently with git worktrees
- Auto PR Creation: Create pull requests automatically
- Cost Tracking: Monitor token usage and costs
Supported AI Engines
| Engine | Flag | CLI |
|---|---|---|
| Claude Code | --claude (default) |
claude |
| OpenCode | --opencode |
opencode |
| Cursor Agent | --cursor |
agent |
| Codex CLI | --codex |
codex |
| Qwen-Code | --qwen |
qwen |
| Factory Droid | --droid |
droid |
Configuration
Initialize a .ralphy/ directory with project config:
ralphy --init
This auto-detects:
- Project name, language, framework
- Test, lint, and build commands
Add rules the AI must follow:
ralphy --add-rule "Always use TypeScript strict mode"
PRD Formats
Markdown:
- [ ] Implement user authentication
- [ ] Add payment processing
- [x] Set up database (completed)
YAML:
tasks:
- title: "Implement auth"
parallel_group: 1
- title: "Add payments"
parallel_group: 1
GitHub Issues:
ralphy --github owner/repo --github-label enhancement
Options
ralphy --parallel # Run tasks in parallel
ralphy --max-parallel 5 # Concurrent agents (default: 3)
ralphy --branch-per-task # Create branch per task
ralphy --create-pr # Create PRs automatically
ralphy --no-tests # Skip tests
ralphy --no-lint # Skip linting
ralphy --fast # Skip tests and linting
ralphy --max-iterations 10 # Stop after N tasks
ralphy --dry-run # Preview without executing
ralphy --verbose # Debug output
Requirements
- Python 3.10+
- git
- At least one AI CLI installed (claude, opencode, etc.)
License
MIT
Metadata
Release files for ralphy 4.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| ralphy-4.0.1.tar.gz | 32.8 kB | Without provenance | Details for ralphy-4.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| ralphy-4.0.1-py3-none-any.whl | Python 3 | none | any | Details for ralphy-4.0.1-py3-none-any.whl |
Total release size: 70.3 kB
Release files / ralphy-4.0.1.tar.gz
| Download URL | ralphy-4.0.1.tar.gz |
|---|---|
| Size | 32.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0e4b1f78420fd9bf607813f866a2ae3688273741ec19dbb09bd382f3186db04e
|
|
BLAKE2b-256 checksum How to use checksums |
ecb621a538d4ec8460efbb0b599673df19a3dd9fe69f26bbed60e5d918719215
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.16
|
Release files / ralphy-4.0.1-py3-none-any.whl
| Download URL | ralphy-4.0.1-py3-none-any.whl |
|---|---|
| Size | 37.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9debe4fa52ba937aa83f6559c56d3a71a4ad7d379f48613391a11c3ab4610612
|
|
BLAKE2b-256 checksum How to use checksums |
928ae7039c1c1ba19cd1b4282dabb4b46f80e96865eec23566e2c1ad4b0931ee
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.16
|
Security for ralphy 4.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.