mcp-server-builder
Skill1 starsChanged 2 months agoMITView on GitHub
Scaffold MCP servers from OpenAPI specs. TRIGGER when: user asks to build an MCP server, convert an OpenAPI/Swagger spec to MCP tools, generate MCP tool definitions, or scaffold a FastMCP/TypeScript MCP project. DO NOT TRIGGER when: user is configuring an existing MCP server, writing MCP clients, or working with Claude API directly (use claude-api skill).
Files
MCP Server Builder
Generate Model Context Protocol servers from OpenAPI specifications. Supports Python (FastMCP) and TypeScript targets.
What You Get
- Working MCP server directory scaffolded from an OpenAPI spec
- Typed tool definitions, auth wrappers, confirmation gates, and test stubs
Workflow
- Parse -- Read the OpenAPI spec (JSON or YAML), extract paths, operations, schemas, and auth requirements.
- Generate -- Map each operation to an MCP tool definition with typed input schemas. Apply naming conventions (verb_noun, snake_case).
- Secure -- Add auth wrappers, host allowlists, confirmation gates for destructive operations, and structured error payloads.
- Validate -- Run the manifest through lint checks: duplicate names, missing descriptions, invalid schemas, naming hygiene.
- Test -- Unit tests for schema transforms, contract tests for manifest snapshots, integration tests against a staging API.
Quality Gates (before publishing)
- Every tool has a non-empty description
- No duplicate tool names
- All destructive operations require confirmation input
- Secrets sourced from env vars only (none in schemas or defaults)
- Host allowlist is explicit (no wildcards unless justified)
- Manifest passes strict validation (
validation.md) - Unit + contract tests pass
- Integration test against at least one live endpoint
Top Pitfalls
| Mistake | Fix |
| Leaking API keys in tool schemas | Use env vars; never put secrets in inputSchema defaults |
Generic tool names (get_data) | Use API-specific prefixes (github_list_repos) |
| Missing error structure | Return {error: string, code: number}, not raw strings |
| Huge parameter objects | Flatten or split; MCP tools work best with focused inputs |
| No confirmation on DELETE | Add confirm: true input for destructive operations |
Reading Guide
| Topic | File |
| OpenAPI-to-MCP mapping, scaffold templates | scaffolding.md |
| Auth patterns, safety, error design | auth-and-safety.md |
| Manifest validation and CI checks | validation.md |
| Testing strategy (unit/contract/integration) | testing.md |
Other versions15 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 15
- OneWave-AI/claude-skills★ 378same name
- timothywarner-org/context-engineering★ 27same name
- Hassaan146/claude-skills★ 18same name
- yeaight7/agent-powerups★ 6same name
- Cure-Consulting-Group/ProductEngineeringSkills★ 3same name
- techsaurabh-394/claude-skills★ 3same name
- csiddhant796-blip/claude-skills-collection★ 2same name
- Hussain-Elfalah/Agentic-Skills★ 2same name
- codecharmhq/claude-code-skills★ 1same name
- mahiunfiltered/claude-code-skills★ 1same name
5 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository61 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- adversarial-reviewerSkill
- agent-designerSkill
- architectSkill
- autoresearchSkill
- blockscoutSkill
- ci-cd-pipeline-builderSkill
- claude-apiSkill
- codebase-onboardingSkill
- code-reviewSkill
- coingeckoSkill
- database-designerSkill
- dependency-auditorSkill
- design-uxSkill
- digestSkill
- droo-stackSkill
- env-secrets-managerSkill
- ethskillsSkill
- ffmpeg-asmSkill
- focused-fixSkill
- frontend-slop-auditSkill
- git-guardrailsSkill
- git-worktree-managerSkill
- humanizerSkill
- interface-designerSkill
- llm-cost-optimizerSkill
- local-browser-inspectorSkill
- native-codeSkill
- nixSkill
- noirSkill
- observability-designerSkill
- patchbotSkill
- performance-profilerSkill
- playwrightSkill
- polymathSkill
- prd-to-planSkill
- prepperSkill
- property-testingSkill
- qaSkill
- rag-architectSkill
- raxol-paymentsSkill
- raxolSkill
- raxol-symphonySkill
- recallSkill
- refactoring-strategySkill
- regenSkill
- releaseSkill
- security-auditorSkill
- self-improving-agentSkill
- sentinelSkill
- signozSkill
- skill-creatorSkill
- solidity-auditorSkill
- tddSkill
- tech-debt-trackerSkill
- ubiquitous-languageSkill
- watchdogSkill
- web-asset-generatorSkill
- writing-voiceSkill
- The first 60 shown.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.