SpecKit MCP
Structured spec management directly in the AI workflow reduces the gap between requirements and implementation.
Specification Management System MCP server for creating, validating, and exporting specs (PRD, RFC, ADR, User Story, API Spec) in YAML/JSON storage with export to Claude Code prompt format.
- Create structured specifications for product requirements and designs
- Validate and export specs in formats Claude can reference later
- Maintain a living library of architectural decision records
Structured spec management directly in the AI workflow reduces the gap between requirements and implementation. ADR templates in particular address a common enterprise request for documenting architectural decisions.
Product managers, architects, and senior engineers who want to maintain structured specifications (PRDs, RFCs, ADRs) accessible to Claude during development.
https://www.npmjs.com/package/@fast-kit/spec-kit
By cphiep96
How to Get It
Per the README: build from the fast-kit monorepo (pnpm install, then pnpm build in implementations/mcp-servers/spec-kit), then add to your Claude config: {"mcpServers": {"spec-kit": {"command": "node", "args": ["/path/to/fast-kit/implementations/mcp-servers/spec-kit/dist/index.js"]}}}
Tip: Paste this into a Claude Code conversation. Verify command matches your Claude Code version.
Once it’s connected, paste this into Claude:
Help me create structured specifications for product requirements and designs
Trust Signals Reviewed
Not yet scanned. Learn how to check this tool yourself.
Data & Access
Reviewer notes
Reviewed review. These are observations, not a security certification.
audit 2026-07-05 stale-eval refresh: Scored from trust signals (evidence-eval-v1): 0 GitHub stars; contributors unknown; commit recency unknown; license MIT.
Things to check
- Requires familiarity with YAML/JSON schema structure and manual spec authoring; no automatic generation from existing codebases or reverse-engineering from running systems.
How to evaluate tools before deploying →
Data shown here comes from public APIs and automated scanning. Reviewer notes reflect one person's experience. This is not a security certification or legal recommendation. Always evaluate tools according to your own organization's policies.
Evaluation
audit 2026-07-05 stale-eval refresh: Scored from trust signals (evidence-eval-v1): 0 GitHub stars; contributors unknown; commit recency unknown; license MIT.