openai-knowledge

Fetches authoritative OpenAI API documentation via the OpenAI Developer Docs MCP server.

Updated May 23, 2026
One-click install
npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill openai-knowledge-kiranimmadi2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: openai-knowledge
Source: https://github.com/kiranimmadi2/promptforge-ai/tree/main/openai-agents-python/.agents/skills/openai-knowledge
Command: npx skills add https://github.com/kiranimmadi2/promptforge-ai --skill openai-knowledge-kiranimmadi2

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When working with the OpenAI API (Responses API, Realtime API, tools, streaming, auth, rate limits, MCP), answers based on memory can be outdated or wrong. This Skill retrieves exact, up-to-date OpenAI documentation so answers are grounded in official sources instead of guesses. ## Core Features & Use Cases - Docs Search and Fetch: Uses the openaiDeveloperDocs MCP server tools to search OpenAI docs and fetch exact markdown pages, optionally by anchor. - API Schema Lookup: Retrieves the OpenAPI spec and endpoint listings to answer questions about parameters, fields, and limits precisely. - Setup Guidance: If the MCP server is not configured, it provides CLI (codex mcp add) or config.toml setup instructions without modifying config unprompted. - Use Case: You are unsure of the exact parameters for the Responses API streaming mode. The Skill searches the official docs, fetches the relevant page, and answers with quoted, accurate details. ## Quick Start Ask the assistant to look up the exact OpenAI documentation for your API question, for example the streaming parameters of the Responses API.

Frequently Asked Questions about openai-knowledge

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

FAQPage Schema
How do I get up-to-date OpenAI API documentation in Codex?

Enable the openaiDeveloperDocs MCP server, then use its search and fetch tools to pull exact documentation pages. Run codex mcp add openaiDeveloperDocs --url https://developers.openai.com/mcp and restart the session.

How to look up OpenAI Responses API parameters and schemas?

Use the MCP tools get_openapi_spec and list_api_endpoints to retrieve the official OpenAPI specification. This gives exact endpoint schemas, parameter names, and defaults instead of relying on memory.

Does this work without the OpenAI Developer Docs MCP server configured?

Without the MCP server, the Skill cannot fetch live documentation. It instead guides you through setup via the codex mcp add CLI command or by editing ~/.codex/config.toml, then asks you to restart the session.

What OpenAI platform topics does the documentation cover?

It covers the Responses API, Realtime API, tools, streaming, authentication, models, rate limits, and MCP. The Skill searches and fetches the relevant official pages for any of these topics.

Why are my OpenAI API answers outdated or incorrect?

Model knowledge can lag behind API changes, leading to invented flags, fields, or limits. Fetching the current documentation through the MCP server grounds answers in the exact published text.