LMCP is a free Mac installer that exposes Mail, Calendar, Slack, Teams, browser, and other local apps as MCP tools for Claude Code, Cursor, Codex, and Claude Desktop—often reading local caches instead of OAuth/cloud APIs. Install via local-mcp.com or npx -y local-mcp@latest setup; recipes save multi-step workflows for scheduled agent runs.
Claude Code 2.1.287 turns on Mods by default: small JS/TS plugins that hook tool calls, prompts, and UI so you can rewrite behavior or draw custom panes—Anthropic already uses them for /diff and AGENTS.md. Install via /plugin ; they run with full machine access, so only load mods you trust…
Cloudeval’s CLI ships a local stdio MCP ( npx y @ganakailabs/cloudeval cli mcp serve toolset readonly ) so Cursor, Claude Code, and Codex can read project architecture graphs plus saved cost and Well Architected reports—scoped keys, no deploy or Azure changes. It’s review context alongside Microsoft’s live Azure MCP Server, not a replacement…
OpenAI hosts a public streamable HTTP MCP at https://developers.openai.com/mcp so Codex, Cursor, Claude Code, and Copilot Agent can search and read OpenAI developer docs in session—docs only, no API calls on your behalf. Add it with codex mcp add , Cursor’s mcp.json , or claude mcp add transport http…
Cloudflare released Clef (27B) and Clef flash (9B)—Jev compatible decision models that return typed yes/no, choice, or score answers with probabilities for agent routing. Apache 2.0 weights are on Hugging Face; hosted inference runs on Workers AI, with an RL fine tuning path for domain specific classifiers…
ssapable’s MIT MCP (plus Claude Code plugin) stores approve/edit/reject decisions in local PGlite+pgvector and injects the closest past corrections before the next draft—so style fixes don’t die with the chat. Install via /plugin or claude mcp add ; hooks cover SessionStart and UserPromptSubmit when tool calls alone aren’t enough…
Spotted something we missed? Start a thread.
