chat_with_vision
analyze local or remote images with a Grok vision model.
How to use it
Project documentation lists chat_with_vision for the Grok MCP MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), then check which tools your installed version makes available. Tool availability can depend on configuration and credentials. A server handshake does not verify this tool’s behavior. See the full listing for setup details.
Install Grok MCP
claude mcp add grok-mcp -e XAI_API_KEY=your_api_key_here -- uv run --directory /path/to/Grok-MCP python main.pyOther tools in Grok MCP (21)
standard chat completion with optional persistent history and multi-agent support.
chat grounded on previously uploaded documents.
delete a session's local history file.
solve tasks by running Python in a sandbox.
delete a file by ID.
delete a stored response by ID.
extend an existing generated video with a follow-up prompt.
create or edit images with Grok Imagine (multi-reference editing supported).
text-to-video, image-to-video, or video editing with Grok Imagine.
get a session's full transcript.
fetch file metadata by ID.
download file content as text.
unified agent that mixes files, images, web search, X search, and code execution.
list saved sessions in chats/.
list uploaded files with sorting.
list all Grok language and image models with live pricing.
fetch a stored response by ID.
continue a server-side stored conversation via responseid.
upload a local document.
autonomous web research with domain filters and citations.
autonomous search over X (Twitter) posts, with handle and date filters.