There should be a way to support MCP servers (both stdio based and http/https based) to use in the internal agent. we only expect local MCP servers, since this is a local-AI focused project. This is only relevant for the local AI chat, not for the external server endpoint, as the server endpoint expects outside agents to manage their own mcp setups.
this should only support .mcp.json, so it is about project-local mcp servers for now, in the case where projects bring predefined mcp servers.
There should be a way to support MCP servers (both stdio based and http/https based) to use in the internal agent. we only expect local MCP servers, since this is a local-AI focused project. This is only relevant for the local AI chat, not for the external server endpoint, as the server endpoint expects outside agents to manage their own mcp setups.
this should only support .mcp.json, so it is about project-local mcp servers for now, in the case where projects bring predefined mcp servers.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
There should be a way to support MCP servers (both stdio based and http/https based) to use in the internal agent. we only expect local MCP servers, since this is a local-AI focused project. This is only relevant for the local AI chat, not for the external server endpoint, as the server endpoint expects outside agents to manage their own mcp setups.
this should only support .mcp.json, so it is about project-local mcp servers for now, in the case where projects bring predefined mcp servers.
while it might become relevant at some point, right now I want to go with internal tools and skills and nothing else.