Our review
Starts an autonomous development loop that executes PRD user stories with smart failure handling, skill-based verification, and checkpointing.
Strengths
- Fully autonomous execution without human oversight after launch
- Robust failure management with circuit breaker and error injection
- Post-step verification using security and performance skills
- Resumable from checkpoint in case of interruption
Limitations
- Requires a pre-defined and structured PRD plan
- Security modes may restrict necessary actions
- May be inefficient for very complex projects without customization
Use this skill when you have a clear PRD plan and want to execute it autonomously with monitoring and resume capability.
Do not use it if the PRD plan is poorly defined, if continuous human supervision is needed, or if required actions exceed the permissions of the chosen security mode.
Security analysis
CautionThe skill orchestrates an autonomous development loop using Bash and code modification tools. While it includes safety measures like circuit breakers and security modes, the inherent power and autonomy pose significant risk if misused or if the PRD contains malicious instructions. Not explicitly destructive, but requires caution.
- •Uses Bash and Write/Edit tools autonomously, which could lead to destructive actions if not properly constrained.
- •Supports 'yolo' security mode that grants full permissions, increasing risk of unintended damage.
- •Autonomous loop may execute arbitrary code based on PRD stories, potentially leading to security breaches.
Examples
/ralph-ultra:ralph-run/ralph-ultra:ralph-run --max-iterations 5 --security sandbox/ralph-ultra:ralph-run --resumename: ralph-run description: Start the Ralph Ultra autonomous development loop. Iterates through PRD stories with smart failure handling, skill-based verification, and checkpointing. Use when ready to execute a PRD plan autonomously. allowed-tools: Read, Write, Edit, Bash, Grep, Glob
Ralph Ultra Autonomous Loop
Start the smart adaptive development loop that iterates through PRD stories.
What this does
- Pre-flight checks — Runs environment-doctor and deep-codebase-navigator
- Story selection — Picks next story respecting dependencies and failure history
- Prompt construction — Builds context from CLAUDE.md, skills, progress, and error history
- Execution — Invokes Claude/AMP with appropriate security flags
- Verification — Runs post-iteration skills (security-auditor, performance-guardian)
- Failure handling — Circuit breaker at 3 failures, error injection for self-correction
- Checkpointing — Saves state after each iteration for resumability
Usage
/ralph-ultra:ralph-run [--max-iterations N] [--security sandbox|standard|yolo] [--tool claude|amp] [--resume]
Options
| Option | Default | Description |
|--------|---------|-------------|
| --max-iterations | 10 | Maximum loop iterations |
| --security | standard | Security mode (sandbox/standard/yolo) |
| --tool | claude | AI tool to use |
| --resume | false | Resume from last checkpoint |
Loop Flow
Pre-flight Skills → Pick Story → Build Prompt → Execute →
↓ Success: Mark done, run post-iteration skills → Next story
↓ Failure: Record error, inject context → Retry (max 3) → Block story
Security Modes
- sandbox: Read-only analysis, no file modifications
- standard: Allowlist-based, only permitted commands
- yolo: Full permissions (requires explicit confirmation)
Monitoring
While the loop runs:
- View live dashboard:
/ralph-ultra:ralph-dashboard - Check status:
/ralph-ultra:ralph-status - Session logs:
.ralph-ultra/sessions/current/log.jsonl
Next.js App Router Expert
Development
A skill that turns Claude into a Next.js App Router expert.
README Generator
Development
Creates professional and comprehensive README.md files for your projects.
API Documentation Writer
Development
Generates comprehensive API documentation in OpenAPI/Swagger format.