{
  "markdown": "<p align=\"center\">\n  <img src=\"assets/logo.png\" alt=\"Contexta\" width=\"320\" />\n</p>\n\n<h1 align=\"center\">Contexta MCP</h1>\n\n<p align=\"center\">\n  <b>Persistent memory & knowledge graph for your AI assistant.</b><br/>\n  Keyword + vector + graph search in a single lookup.\n</p>\n\n<p align=\"center\">\n  <a href=\"https://smithery.ai/servers/ai-jcce/mcp-contexta\">\n    <img src=\"https://img.shields.io/badge/Smithery-published-2dd4bf?style=flat-square\" alt=\"Smithery\" />\n  </a>\n  <a href=\"https://contexta.cc\">\n    <img src=\"https://img.shields.io/badge/contexta.cc-website-07080d?style=flat-square\" alt=\"Website\" />\n  </a>\n  <a href=\"LICENSE\">\n    <img src=\"https://img.shields.io/badge/license-MIT-blue?style=flat-square\" alt=\"License\" />\n  </a>\n</p>\n\n---\n\n## What is Contexta\n\nContexta is a memory MCP server that remembers like a human does — by meaning, by words, and by relationships.\n\nMost memory tools give you one flavor of recall: embedding search, a raw log, or a graph you pay extra for. Contexta fuses all three into a single query. Your agent can ask _\"what did we decide about pricing with Acme last quarter?\"_ and get back the right meeting, the right decision, and the people connected to it — in one round-trip.\n\nUnder the hood, Contexta indexes your notes, messages, and documents as **typed entities** (people, projects, decisions, meetings, tasks) and links them into a knowledge graph automatically. Every retrieval combines:\n\n- **Keyword search** — exact phrases, names, IDs\n- **Vector search** — semantic meaning across languages\n- **Graph traversal** — follow relationships between entities\n\n## Why Contexta\n\n- **Typed entities out of the box** — people, projects, decisions, meetings, tasks are first-class, not blobs of text\n- **Knowledge graph included** — no paid tier gate on the graph layer\n- **Triple retrieval in one call** — no need to stitch together three different memory APIs\n- **30-second setup** — install via Smithery, sign in with OAuth, done\n- **Per-user OAuth** — your memory is private; the server authenticates every request individually\n- **Multilingual** — English and Russian equally supported\n\n## Use cases\n\n- Persistent memory for Claude, ChatGPT, and Cursor across sessions\n- Long-term context for AI agents working on ongoing projects\n- Personal knowledge base that actually understands who and what you mean\n- Research and meeting recall without manually tagging anything\n\n## Install\n\n### Via Smithery (recommended)\n\nOne-click install from the [Smithery listing](https://smithery.ai/servers/ai-jcce/mcp-contexta) — Smithery handles the OAuth flow and proxies connections through its gateway.\n\n### Manual — Claude Desktop\n\nAdd to your Claude Desktop config (`~/Library/Application Support/Claude/claude_desktop_config.json` on macOS, `%APPDATA%\\Claude\\claude_desktop_config.json` on Windows):\n\n```json\n{\n  \"mcpServers\": {\n    \"contexta\": {\n      \"url\": \"https://mcp.contexta.cc/mcp\"\n    }\n  }\n}\n```\n\nClaude will walk you through the OAuth sign-in on first use. See [examples/claude-desktop.json](examples/claude-desktop.json).\n\n### Manual — Cursor\n\nAdd a new MCP server in Cursor → Settings → MCP:\n\n- **Name:** Contexta\n- **Transport:** HTTP\n- **URL:** `https://mcp.contexta.cc/mcp`\n\nSee [examples/cursor.json](examples/cursor.json).\n\n### Manual — ChatGPT\n\nIn ChatGPT → Settings → Connectors → Add custom connector → paste `https://mcp.contexta.cc/mcp`. See [examples/chatgpt.md](examples/chatgpt.md) for full instructions.\n\n## Authentication\n\nContexta uses **OAuth 2.0 with dynamic client registration** (RFC 7591) and PKCE. On first connection, your MCP client opens a browser window, you sign in with your Contexta account, and the client receives a per-user access token. Tokens are refreshed automatically.\n\nNo API keys, no shared credentials — every request runs as a specific user with that user's private memory.\n\n## Tools exposed\n\nContexta MCP exposes the following tools to your AI client (non-exhaustive):\n\n- `search` — keyword + vector + graph search across your memory\n- `read_page` — fetch a note or document by ID\n- `create_page` — save a new memory\n- `update_page` — edit an existing memory\n- `graph_search` — traverse the knowledge graph\n- `list_tasks` / `create_task` / `update_task` — task management\n- `list_workspaces` — available workspaces\n- and more — the full tool list is discovered automatically by your MCP client.\n\n## Links\n\n- **Website:** [contexta.cc](https://contexta.cc)\n- **App:** [app.contexta.cc](https://app.contexta.cc)\n- **Smithery:** [smithery.ai/servers/ai-jcce/mcp-contexta](https://smithery.ai/servers/ai-jcce/mcp-contexta)\n- **MCP endpoint:** `https://mcp.contexta.cc/mcp`\n\n## License\n\nMIT — see [LICENSE](LICENSE).\n\nThis repository is a public showcase for the hosted Contexta MCP service. The service itself is operated by the Contexta team at `mcp.contexta.cc`.\n",
  "bytes": 4852,
  "sha": "8e8139466e0eb5dccdf353af8ade2dd269b8b0192ea250d58e7f5a3fb82d19cf",
  "repo_slug": "teleengco/contexta-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_cc_contexta_contexta_mcp_0c509e1f/readme"
}