Is your feature request related to a problem? Please describe.
Currently, integrating Stream Video/Chat into complex existing codebases can be time-consuming. While the documentation is extensive, standard AI coding assistants (Copilot, ChatGPT, etc.) often struggle to provide accurate, context-aware solutions. They frequently:
Redirect to documentation URLs rather than providing specific code fixes.
Lose track of SDK version differences (e.g., mixing up Video vs. Chat syntax).
Fail to account for the user's specific state management (React/Next.js) or backend architecture when fetching docs.
Describe the solution you'd like
I would like to see GetStream provide official support for the Model Context Protocol (MCP) and LLM-optimized documentation. Specifically:
Official GetStream MCP Server: A pre-built MCP server that IDEs like Cursor, Claude Desktop, and VS Code (via Copilot) can use to index the latest Stream documentation, API references, and migration guides. This would allow the LLM to have "live" access to the Stream ecosystem while reading my local code.
LLM-Friendly Docs (llms.txt / .md): Following the lead of projects like Better Auth, Strapi, and Stripe, provide a /docs.md or a dedicated LLM-optimized page. This single-page, markdown-heavy resource allows AI tools to scrape and "understand" the library's entire logic without navigating 50+ nested sub-pages.
Context-Aware Integration Skills: Tools that can "read" my current implementation and suggest the exact Stream component or hook (e.g., CallContent, StreamCall) needed to fix a specific bug.
Describe alternatives you've considered
Manually copying and pasting large chunks of documentation into Claude/GPT-4, which quickly hits token limits and loses formatting.
Using generic web-search agents, which often pull outdated information from older SDK versions.
Additional context
Providing these "AI-native" documentation tools would significantly lower the barrier to entry for new developers and drastically reduce debugging time for existing customers.
Example of great LLM docs: Better Auth Introduction
Reference for MCP: Model Context Protocol
Is your feature request related to a problem? Please describe.
Currently, integrating Stream Video/Chat into complex existing codebases can be time-consuming. While the documentation is extensive, standard AI coding assistants (Copilot, ChatGPT, etc.) often struggle to provide accurate, context-aware solutions. They frequently:
Redirect to documentation URLs rather than providing specific code fixes.
Lose track of SDK version differences (e.g., mixing up Video vs. Chat syntax).
Fail to account for the user's specific state management (React/Next.js) or backend architecture when fetching docs.
Describe the solution you'd like
I would like to see GetStream provide official support for the Model Context Protocol (MCP) and LLM-optimized documentation. Specifically:
Official GetStream MCP Server: A pre-built MCP server that IDEs like Cursor, Claude Desktop, and VS Code (via Copilot) can use to index the latest Stream documentation, API references, and migration guides. This would allow the LLM to have "live" access to the Stream ecosystem while reading my local code.
LLM-Friendly Docs (llms.txt / .md): Following the lead of projects like Better Auth, Strapi, and Stripe, provide a /docs.md or a dedicated LLM-optimized page. This single-page, markdown-heavy resource allows AI tools to scrape and "understand" the library's entire logic without navigating 50+ nested sub-pages.
Context-Aware Integration Skills: Tools that can "read" my current implementation and suggest the exact Stream component or hook (e.g., CallContent, StreamCall) needed to fix a specific bug.
Describe alternatives you've considered
Manually copying and pasting large chunks of documentation into Claude/GPT-4, which quickly hits token limits and loses formatting.
Using generic web-search agents, which often pull outdated information from older SDK versions.
Additional context
Providing these "AI-native" documentation tools would significantly lower the barrier to entry for new developers and drastically reduce debugging time for existing customers.
Example of great LLM docs: Better Auth Introduction
Reference for MCP: Model Context Protocol