.claude/skills/prd-check/SKILL.md
Run PRD validation to check story quality, test coverage, and structure.
npx skillsauth add allierays/agentic-loop .claude/skills/prd-checkInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Run PRD validation on demand to check story quality, test coverage, and structure before starting the autonomous loop. Cross-references against learned lessons and past failure patterns.
When the user runs /prd-check, validate their PRD file.
ls -la .ralph/prd.json 2>/dev/null || echo "NOT_FOUND"
If no PRD exists, tell the user:
No PRD found at
.ralph/prd.json. Generate one first with/prd.
STOP if no PRD found.
Read the project's accumulated knowledge:
Read lessons:
cat .ralph/lessons.json 2>/dev/null
Read suggested lessons (last 50 lines — file can be huge):
tail -50 .ralph/suggested-lessons.txt 2>/dev/null
Read recent progress for failure patterns:
tail -100 .ralph/progress.txt 2>/dev/null
npx ralph prd-check --dry-run 2>&1
Present any structural issues found.
Read .ralph/prd.json and for each story, check if any lesson applies:
For each applicable lesson, verify the story's acceptanceCriteria, constraints,
notes, or testSteps reflect the pattern. Flag stories that should account for
a lesson but don't.
Examples:
Scan suggested-lessons.txt for patterns relevant to the current PRD's feature area.
Flag any recurring failure patterns that the PRD's stories don't address.
Summarize all findings in categories:
Structural Issues (from prd-check):
- [list]
Lesson Conflicts (stories that don't account for known patterns):
- TASK-003: Missing lesson "bcrypt cost 10+" in auth story
- TASK-005: Missing lesson "data-testid attributes" in frontend story
Suggested Improvements (from past learnings):
- [relevant patterns from suggested-lessons.txt]
Ask: "Would you like me to fix these issues in the PRD?"
STOP and wait for user response.
If yes, update .ralph/prd.json:
constraints or acceptanceCriteriaralph run startup--dry-run skips auto-fix so you have control over what changes.ralph/checks/prd/ are also evaluatedralph run to catch and fix issues interactivelytools
Show complete reference for all agentic-loop commands (slash commands, Ralph CLI, vibe CLI).
data-ai
Quick reference cheatsheet for all agentic-loop commands including Ralph.
development
Run a comprehensive code quality check looking for common patterns that AI coding agents introduce.
development
Take an interactive tour of agentic-loop - the system for going from idea to shipped code with AI.