oas-api-spec-generator

Generates OpenAPI 3.2.0 specifications for third-party REST APIs from official documentation.

Updated Feb 23, 2026
One-click install
npx skills add https://github.com/deepparser/skills --skill oas-api-spec-generator-deepparser
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oas-api-spec-generator
Source: https://github.com/deepparser/skills/tree/main/oas-api-spec-generator
Command: npx skills add https://github.com/deepparser/skills --skill oas-api-spec-generator-deepparser

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writing OpenAPI specifications for third-party APIs by hand is slow and error-prone, especially when providers publish only prose documentation or outdated spec versions. This Skill automates the creation of validated OpenAPI 3.2.0 documents for external APIs you integrate with. ## Core Features & Use Cases - Provider-aware generation: Built-in profiles for 15+ providers (OpenAI, Anthropic, Google, Microsoft Graph, Stripe, GitHub, Slack, AWS, Twilio, and more) with correct auth schemes, base URLs, and pagination patterns. - OAS 3.2.0 features: Structured tags, reusable mediaTypes, webhooks, and native streaming support via text/event-stream and application/x-ndjson with itemSchema. - Official spec upgrading: Fetches a provider's published OpenAPI spec when available and upgrades it to 3.2.0 instead of writing from scratch. - Validation and multi-file output: Checks specs against 11 mandatory rules and 10 best practices, and splits large specs (50+ endpoints) into modular path and component files. - Use Case: You are building an SDK or API gateway integration for the Slack Web API. Ask the Skill to generate the spec, and it produces a validated specs/slack/openapi.yaml with bot-token auth, cursor pagination, and standard error responses. ## Quick Start Ask the assistant to generate an OpenAPI 3.2.0 spec for the provider you are integrating with, for example by saying create an OAS spec for the Stripe API.

Frequently Asked Questions about oas-api-spec-generator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an OpenAPI spec for a third-party API?

Provide the provider name and the endpoints you need, and the Skill fetches the provider's official documentation or published spec. It then generates a validated OpenAPI 3.2.0 YAML file saved to specs/{provider}/openapi.yaml.

Which API providers are supported for OpenAPI spec generation?

Built-in profiles cover OpenAI, Anthropic, Google AI, Microsoft Graph, Stripe, GitHub, Slack, AWS, Cloudflare, Vercel, Twilio, SendGrid, and more. Any other REST API with accessible documentation can also be processed.

Does the OpenAPI 3.2.0 spec support streaming responses like SSE?

Yes, streaming is modeled natively using text/event-stream or application/x-ndjson media types with an itemSchema describing each streamed chunk. This is used for AI APIs such as OpenAI and Anthropic chat completions.

Can it upgrade an existing OpenAPI 3.0 spec to 3.2.0?

Yes, when a provider publishes an official spec, the Skill fetches it and upgrades it to OAS 3.2.0 rather than writing from scratch. It also adds missing features like structured tags and streaming definitions during the upgrade.

How are large API specs with many endpoints organized?

Specs with 50 or more endpoints are split into multiple files: an entry openapi.yaml plus separate directories for paths, schemas, parameters, responses, and security schemes. Cross-file references use standard $ref pointers.

What validation rules does the generated OpenAPI spec follow?

Mandatory checks include the exact 3.2.0 version field, valid $ref targets, unique operationIds, defined security schemes, lowercase HTTP methods, and string status codes. Additional warnings cover descriptions, error schemas, pagination, and rate-limit headers.