Agent skill · AI & Agents

ralph

Convert PRDs to prd.json format and run the Ralph autonomous agent system. Use when converting PRDs to Ralph format OR running Ralph to execute user stories. Triggers on: convert this prd, turn this into ralph format, create prd.json, ralph json, run ralph, start ralph, execute ralph.

majiayu000github.com/majiayu000GitHub ↗
claude-codeMIT
Install
npx skills add majiayu000/claude-skill-registry --skill ralph-gannonh-kata-agents --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 2
SKILL.md size: 8 KB
Bundled scripts: none
Path: skills/agent/ralph-gannonh-kata-agents/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 534
Language: HTML

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# Ralph - Autonomous Agent System Ralph is an autonomous agent that executes user stories from a `prd.json` file. This skill covers both: 1. **Converting PRDs** to the prd.json format Ralph requires 2. **Running Ralph** using the `./scripts/ralph.sh` script --- ## Running Ralph Execute Ralph using the script at `./scripts/ralph.sh`: ```bash # Run with default settings (amp tool, 10 iterations) ./scripts/ralph.sh # Run with Claude Code instead of amp ./scripts/ralph.sh --tool claude # Run with custom max iterations ./scripts/ralph.sh 20 # Combine options ./scripts/ralph.sh --tool claude 15 ``` **Options:** - `--tool amp|claude` - Choose which AI tool to use (default: amp) - `[max_iterations]` - Maximum iteration count (default: 10) **How it works:** 1. Reads `prd.json` from the script directory 2. Spawns a fresh AI instance per iteration (no memory between iterations) 3. Each iteration works on the next incomplete user story 4. Archives previous runs when branch name changes 5. Exits when all stories pass or max iterations reached --- ## PRD Conversion Converts existing PRDs to the prd.json format that Ralph uses for autonomous execution. --- ## The Job Take a PRD (markdown file or

What's inside
Steps it walks through
  1. Running Ralph
  2. PRD Conversion
  3. The Job
  4. Output Format
  5. Story Size: The Number One Rule
  6. Right-sized stories:
  7. Too big (split these):
  8. Story Ordering: Dependencies First
  9. Acceptance Criteria: Must Be Verifiable
  10. Good criteria (verifiable):
  11. Bad criteria (vague):
  12. Always include as final criterion:
  13. For stories that change UI, also include:
  14. Conversion Rules
Ships with 1 file
  • metadata.json
Commands it runs
Run with default settings (amp tool, 10 iterations)
Run with Claude Code instead of amp
Run with custom max iterations
Combine options
More from claude-skill-registry
All skills →
About this skill
What does the ralph skill do?

Convert PRDs to prd.json format and run the Ralph autonomous agent system. Use when converting PRDs to Ralph format OR running Ralph to execute user stories. Triggers on: convert this prd, turn this into ralph format, create prd.json, ralph json, run ralph, start ralph, execute ralph.

How do I install it?

Run `npx skills add majiayu000/claude-skill-registry --skill ralph-gannonh-kata-agents --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going