agentic-dev-kit
The open-source toolkit behind every exhibit: Claude Code skills, an MCP server, and PRD templates that make agentic building repeatable.
Product
The shipped product: live link, the problem it solves, and the stack it runs on.
The problem
Agentic coding is powerful but ad-hoc. This kit makes it repeatable: a PRD loop, a verify gate, design-sync, spec reconciliation, and project-introspection tools you can drop into any repo.
Stack
Preview
$ agent run --kit ✓ skills · prd · implement · verify → mcp · schema · scaffold · deploy ✓ ready
How I built it
Product requirements
PRD not published for this exhibit yet.
A reusable toolkit extracted while building this portfolio: skills (prd / prd-grill-me / implement / verify / design-sync / reconcile), an MCP server (schema introspection, exhibit scaffolding, deploy status), and PRD templates.
Recent commits
All commitsLive from GitHub, as of Aug 8, 2026
- a1c7336Jun 30, 2026
feat: thread inputs/outputs/edge-cases/seam/tests through the PRD->implement loop (#3)
- 984929aJun 25, 2026
feat: add learnings skill to close the loop with a learn step (#2)
- 2f4fc29Jun 24, 2026
feat: add prd-grill-me skill to harden a draft PRD before implement
- 49926f0Jun 24, 2026
- 15ae82cJun 22, 2026
feat: scaffold the agentic-dev-kit (skills, templates, MCP server)
Metrics
Each metric carries an honesty tag. Verified numbers read solid; targets and placeholders read muted.
Built the same way, every time.
Every exhibit runs the same loop: spec, implement, verify, ship. If you want this kind of work on your team, full-time or freelance, let's talk.