OpenAPI 2 Skill – progressive disclosure of OpenAPI specs for agents
Progressive disclosure beats MCP tool flooding for large OpenAPI specs.

Semantic field inference auto-detects emails and UUIDs for Zod schemas better than quicktype.
Full-stack developers, API integrators
quicktype.io · transform.tools
Progressive disclosure beats MCP tool flooding for large OpenAPI specs.
OpenAPI to pytest generator when Schemathesis and FastAPI already do this.
It extracts focused, executable operations from giant OpenAPI files (the GitHub REST YAML is shown) to shrink context and avoid sidecar adapter sprawl — a pragmatic answer to token bloat and brittle ad-hoc integrations. Useful and concrete: if it actually generates tidy, updateable skill units and runtime hooks it saves a lot of maintenance. That said, the idea competes with existing LangChain/openai-function patterns; the repo will need clear runtime, versioning, and update strategies to feel like more than a nicer converter.
Native binary output beats curl wrappers when you need typed API commands.
Polished UI, but identical tools exist free on CyberChef, in VS Code, and browser DevTools.
MCP-as-CLI cuts tool schema tokens 96-99%, discovered on demand not injected.