llm-mcp-client Lets Simon's llm CLI Talk to Any MCP Server
llm-mcp-client is a new alpha-stage plugin that lets Simon Willison's llm command-line tool connect to and call tools from Model Context Protocol servers.
What it is
llm-mcp-client is an early alpha (0.1a0) tool built to connect the llm CLI to MCP servers, built in response to the newly released Stateless MCP spec.
What it does
It gives the llm command-line tool a way to discover and invoke tools exposed by any MCP server, instead of requiring a dedicated agent framework just to test MCP tool integrations.
Why it matters
It lowers the bar for experimenting with MCP servers from the terminal, an alternative to spinning up a full agent harness just to poke at a server's exposed tools.
How to use it
Install the release from its GitHub tag and point it at an MCP server endpoint to list and call its tools via the llm CLI.
DeepSeek's New Flash Model Costs $0.14/Mtok and Outranks MiniMax M3
DeepSeek-V4-Flash-0731 is a 304 billion parameter open-weight model with enhanced agentic capabilities, priced at $0.14 per million input tokens and $0.27 per million output tokens.
Claude Code 2.1.219 Makes Claude Opus 5 the Default Opus Model
Claude Code 2.1.219 makes Claude Opus 5 the default Opus model with 1M context and fast mode at $10/$50 per Mtok, and adds a sandbox.network.strictAllowlist setting to silently deny non-allowlisted hosts.
Claude Code 2.1.217 Adds Emoji Autocomplete and Fixes an MCP Memory Leak
Claude Code 2.1.217 adds emoji shortcode autocomplete, warns when transcript writes fail instead of losing sessions silently, and fixes a memory leak from truncated MCP tool outputs.