Build MCP servers with Node/TypeScript SDK — tools, resources, prompts, Zod validation, stdio vs Streamable HTTP. Use Context7 or official MCP docs for latest API.
Build MCP servers with Node/TypeScript SDK — tools, resources, prompts, Zod validation, stdio vs Streamable HTTP. Use Context7 or official MCP docs for latest API.
Connect an AI agent to opp_repl via its built-in Model Context Protocol (MCP) server. Two transports — a Unix domain socket reached through the `opp_repl_mcp_bridge` stdio bridge (recommended for local tools like Claude Code, Cursor, VS Code, Windsurf) and a TCP/streamable-HTTP port with bearer-token auth (for remote/legacy clients). Exposes the `execute_python` tool plus opp-repl:// documentation resources. Load when wiring opp_repl into an MCP-capable client.
Build or review a Model Context Protocol (MCP) server that exposes tools, resources, or prompts to AI clients: SDK and transport choice, model-facing tool names, descriptions, and schemas, tool errors, authorization, least privilege, injection defenses, testing with the MCP Inspector, and backward-compatible versioning. Use when creating, extending, or auditing an MCP server. Not for application code that calls a model API directly (use the build-llm-feature skill).
183mo agoReviewMountCreate a free account to mount
Enable Claude to trade, analyze, and manage Polymarket prediction markets with 45 AI-powered tools, real-time monitoring, and enterprise-grade safety features
Operate a running SpacetimeDB database through MCP tools rather than the CLI - list databases, read schemas, run SQL, and call reducers. Use when the client exposes spacetimedb MCP tools and the task is to inspect or change data in a live database.
Add, authenticate, list, and remove MCP (Model Context Protocol) servers — connect any external tool or service that publishes an MCP endpoint to the assistant
Use external MCP services generically from Python - search the supported-service catalog, inspect the user's connections, discover live tool schemas, and call tools on any connection (Notion, Linear, Slack, and the rest of the catalog) without per-service packages.
Add or configure an MCP server in opencode's config (~/.config/opencode/opencode.json or project .opencode/) — both server types: LOCAL/stdio {"type": "local", "command": ["npx", "pkg"], "enabled": true} and REMOTE/HTTP {"type": "remote", "url": "...", "enabled": true}. Use when the user says 'add the X MCP server', 'install/configure MCP', 'connect to <service> MCP', an MCP server appears in docs but isn't wired into opencode, tools from a known MCP server are missing in a session, or a server needs REINSTALLING/RESTORING after its entry was dropped. KEY GOTCHAS: servers load at STARTUP (restart to see new ones); many remote servers (Exa) work FREE without an API key — start with type+url+enabled only; API keys go in a headers block; OAuth servers (Notion, Linear) use "oauth": {} and the user must run 'opencode mcp auth <name>' (the agent cannot complete OAuth); verify JSON validity after editing. Verified examples for Exa/Notion/Jupyter live in the body.
4147d agoReviewMountCreate a free account to mount
MCP Apps integration for json-render. Use when building MCP servers that render interactive UIs in Claude, ChatGPT, Cursor, or VS Code, or when integrating json-render with the Model Context Protocol.
Use when packaging, validating, or distributing a uv-managed Python MCP server (new or existing) as an MCPB bundle (.mcpb) for Claude Desktop or another MCPB host. Triggers include creating an MCPB manifest.json, setting server.type to uv, bundling a uv + pyproject.toml stdio server, running mcpb pack or mcpb validate, preparing one-click install, and exposing API tokens or install-time settings through user_config. Scoped to the MCPB packaging step (manifest_version 0.4+); not for authoring the server's tools or handlers, designing/reviewing/auditing MCP tool/resource/prompt schemas, Node servers, or cloud-only remote HTTP servers.
Backend development guide for the Phoenix AI observability platform (Strawberry GraphQL, SQLAlchemy async, FastAPI). Use this skill when writing or modifying Python server code in the phoenix repo — adding mutations, types, migrations, or tests. Trigger on any backend task touching src/phoenix/server/, src/phoenix/db/, or tests/unit/server/.
Steady8 mentions in 2 weeks
No other project uses it yet
12k2mo agoReviewMountCreate a free account to mount
Plain text files in a repository that tell a coding agent how the project works: commands to run, conventions to follow and things to avoid. CLAUDE.md, AGENTS.md, cursor rules and skills are the common kinds.
CLAUDE.md or AGENTS.md?
CLAUDE.md is read by Claude Code. AGENTS.md is an open format that Codex, Cursor and other agents read. Many projects keep one and point the other at it.
What is a skill?
A folder with a SKILL.md that describes one capability, such as filling PDFs or reviewing code. The agent loads it only when the task calls for it.
Can I search my own team's files too?
Your agents already can, over MCP, limited to the files you're allowed to read. Searching them from this page is coming.