prd-to-sdlc
Skill0 starsChanged 5 months agoMITView on GitHub
Convert a one-page PRD into a full SDLC artifact set — architecture, DB schema, code skeleton, eval harness, and deploy config — by routing through MetaGPT / OpenHands / Promptfoo with PO-grade guardrails baked in (Y-Score readiness, regulated-AI checks, anti-hallucination patterns).
In this repository6 other files it gives its agents.
- agents.mdAGENTS.md
- AGENTS.mdAGENTS.md
- pptx-officialSkill
- y-score-readinessSkill
- cavemanSkill
- productowner-skillSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 58
Anti-hallucination: every architecture claim cites the PRD line that justifies it.
- Very specific: line 59
Compliance-aware: if domain ∈ {bfsi, healthcare}, inject HIPAA/PCI/GDPR/EU-AI-Act checklists into Architect prompt.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
prd-to-sdlc
A reusable Claude Skill that takes a PRD and produces a deployable SDLC bundle.
When to trigger
- User says "turn this PRD into a project"
- User says "scaffold the repo for [PRD path]"
- User invokes
/prd-to-sdlc <prd.md> - A new file lands in
prds/on theproductowner-skillrepo
Inputs
| Field | Required | Notes |
prd_path | ✅ | Markdown PRD with at minimum: problem, users, success metrics, constraints |
domain | optional | bfsi / healthcare / devtools / generic — toggles compliance guardrails |
deploy | optional | cloudflare-pages (default), none |
Stages
1. PO gate — Y-Score readiness
Score the PRD against 7 dimensions: problem clarity, target user, success metric, constraints, compliance, eval plan, rollback. Block if score < 70%. See .claude/skills/y-score-readiness/.
2. Architect
Invoke MetaGPT with PRD → emit architecture.md (Mermaid C4) + db/schema.sql + api/openapi.yaml.
3. Designer
Invoke v0 / Onlook with the user-facing flows from PRD → emit ui/ (React + Tailwind).
4. Engineer
Invoke OpenHands with the spec bundle → emit code/ (working app), runs tests until green.
5. QA
Generate Promptfoo evals/promptfooconfig.yaml from PRD success metrics. Generate Playwright smoke tests from UI flows.
6. DevOps
Emit wrangler.toml + .github/workflows/deploy.yml. If CLOUDFLARE_API_TOKEN set, deploy.
Outputs
outputs/<slug>/
├── PRD.md
├── architecture.md
├── db/schema.sql
├── api/openapi.yaml
├── ui/
├── code/
├── evals/
├── wrangler.toml
└── README.md # auto-generated, attribution to productowner-skill + upstreamGuardrails (always on)
- Anti-hallucination: every architecture claim cites the PRD line that justifies it.
- Compliance-aware: if
domain ∈ {bfsi, healthcare}, inject HIPAA/PCI/GDPR/EU-AI-Act checklists into Architect prompt. - Eval-first: refuse to mark SDLC complete without passing Promptfoo eval gate.
- No client PII: scan all outputs against a PII regex pack before commit.
Upstream credits
MetaGPT · OpenHands · Promptfoo · Onlook · Wrangler. See CREDITS.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

