{
  "markdown": "<p align=\"center\">\n  <img src=\"assets/logo.png\" alt=\"Agent Magnet\" width=\"600\">\n</p>\n\n<p align=\"center\">\n  <img src=\"https://img.shields.io/pypi/v/agent-magnet?label=PyPI&labelColor=111827&color=8B5CF6\" alt=\"PyPI\">\n  <a href=\"https://github.com/helinakdogan/agentmagnet/blob/main/LICENSE\"><img src=\"https://img.shields.io/badge/License-MIT-A855F7\"></a>\n</p>\n\n> Your AI forgets your project every new session. This fixes that.\n\n## What it does\n\n- Learns automatically as you work — no \"remember this\" required\n- Shares one memory across Claude, Cursor, and Codex\n- Lets you forget or mark things done — memory doesn't just pile up\n\n## Quick start\n\n```bash\npip install agent-magnet\nagent-magnet init\n```\n\nThat's the whole install. No Redis, no keys, local by default.\n\n## What it remembers\n\nDecisions, conventions, watch-outs, failed attempts, goals, preferences, and actions taken — one memory, sorted automatically.\n\n## Team memory\n\nShared memory for a team — one person's decision becomes everyone's context. Requires a paid Agent Magnet key from [agentmagnet.app](https://agentmagnet.app).\n\n## Works everywhere\n\nClaude, Cursor, Codex — same memory, local by default. `agent-magnet init` detects and configures Claude Code, Claude Desktop, and Cursor for you.\n\n<details>\n<summary>Manual MCP config (Codex, or any other MCP client)</summary>\n\n```json\n{\n  \"mcpServers\": {\n    \"agent-magnet\": {\n      \"command\": \"agent-magnet-mcp\",\n      \"env\": { \"MAGNET_USER_ID\": \"your_name\" }\n    }\n  }\n}\n```\n</details>\n\n---\n\nMIT licensed. [Contribute](.github/CONTRIBUTING.md) — and a ⭐ if this saved your context window.\n",
  "bytes": 1607,
  "sha": "a953e682e1b8e77fc47c63ed40f8f2de00c4b7c45bb460c7299c297ad6d9c713",
  "repo_slug": "helinakdogan/magnet-gateway",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_helinakdogan_agent_magnet_ff23d5bc/readme"
}