{
  "markdown": "# Amber\r\n\r\n**Long-term memory for AI assistants.**\r\n\r\nAmber is an MCP server that gives any AI assistant persistent, searchable memory across conversations. Your AI remembers preferences, decisions, project context, and personal details - without you doing anything special.\r\n\r\n> Just talk normally. Amber stores what matters and finds it when relevant.\r\n\r\n## Quick Install\r\n\r\nOne command. Works with any MCP-compatible client.\r\n\r\n### Claude Code / Claude Desktop\r\n\r\n```bash\r\nclaude mcp add --transport http --scope user amber https://mcp.ambermem.com\r\n```\r\n\r\n### Cursor\r\n\r\nAdd to `~/.cursor/mcp.json` (or `%USERPROFILE%\\.cursor\\mcp.json` on Windows):\r\n\r\n```json\r\n{\r\n  \"mcpServers\": {\r\n    \"amber\": {\r\n      \"url\": \"https://mcp.ambermem.com\"\r\n    }\r\n  }\r\n}\r\n```\r\n\r\n### ChatGPT\r\n\r\nSettings → Connectors → Create → URL: `https://mcp.ambermem.com`\r\n\r\n### Windsurf\r\n\r\nAdd to `~/.codeium/windsurf/mcp_config.json`:\r\n\r\n```json\r\n{\r\n  \"mcpServers\": {\r\n    \"amber\": {\r\n      \"serverUrl\": \"https://mcp.ambermem.com\"\r\n    }\r\n  }\r\n}\r\n```\r\n\r\n### VS Code (GitHub Copilot)\r\n\r\nAdd to `.vscode/mcp.json`:\r\n\r\n```json\r\n{\r\n  \"servers\": {\r\n    \"amber\": {\r\n      \"type\": \"http\",\r\n      \"url\": \"https://mcp.ambermem.com\"\r\n    }\r\n  }\r\n}\r\n```\r\n\r\n### Any MCP client\r\n\r\nURL: `https://mcp.ambermem.com` | Transport: Streamable HTTP | Auth: OAuth 2.1 (auto-discovered)\r\n\r\n## How It Works\r\n\r\n1. **You talk to your AI normally.** Amber stores important facts in the background.\r\n2. **Next conversation,** your AI searches Amber automatically when context would help.\r\n3. **Memory improves over time.** The more you use it, the better it gets.\r\n\r\nNo configuration. No tagging. No manual organization.\r\n\r\n## What Makes Amber Different\r\n\r\n| Feature | Basic memory servers | Amber |\r\n|---------|---------------------|-------|\r\n| Storage | One embedding per memory | **Multiple semantic variants** per fact |\r\n| Search | Single vector lookup | **Hybrid: vector + keyword + RRF fusion** |\r\n| Queries | Exact match only | **Many phrasings per fact**, matched semantically |\r\n| Input | Stored as-is | **Chunked into atomic facts**, each independently searchable |\r\n| Topics | Manual tags or none | **Auto-grouped**, matched semantically at search time |\r\n| Privacy | Varies by server | **No generative model reads your memories** |\r\n| Time | No temporal awareness | **Natural language time parsing** (\"last week\", \"3 days ago\") |\r\n\r\n## Technical Details\r\n\r\n- **25 MCP tools** (14 memory, 9 account, 2 feedback/notification)\r\n- **Hybrid retrieval pipeline**: vector search + full-text search + Reciprocal Rank Fusion\r\n- **Multi-variant embeddings**: every fact is stored with several paraphrases — at least 5, and a store is rejected below that — which is the main thing that makes it findable later\r\n- **No generative model reads your memories**: your assistant does the chunking, the topics and the phrasings. Memory text is sent only to an *embedding* endpoint (a vector model, not a chat model), so no chat model is ever shown what you store\r\n- **Atomic facts**: a conversation becomes individually searchable facts rather than one blob, each with its own subjects, topics and dates\r\n- **Temporal search**: \"what did I say last week?\" resolves to a real date range rather than a keyword match\r\n- **Automatic topic grouping**: memories are grouped by topic, and a search for \"work\" also finds \"career\" and \"job\"\r\n- **Async processing**: storage completes in the background, never blocking your conversation\r\n\r\n## Pricing\r\n\r\n- **60-day free trial** - no charge, cancel anytime\r\n- **$2.99/month** after trial, via PayPal\r\n- **Cancel instantly** - ask your AI to cancel, or cancel through PayPal directly\r\n- **No lock-in** - export all your data as JSON anytime\r\n\r\n## Privacy\r\n\r\n- No email collected\r\n- No marketing, no spam\r\n- Data isolated per user (separate database)\r\n- PayPal handles all payment info\r\n- Full export + account deletion available\r\n- GDPR compliant (data minimization by design)\r\n\r\n## Architecture\r\n\r\nAmber runs on Cloudflare Workers (zero cold starts, global edge deployment) with Turso databases (one per user, full isolation). LLM processing uses Gemini Flash for chunking/expansion and OpenAI for embeddings.\r\n\r\nFor full technical documentation: [ambermem.com/llms.txt](https://ambermem.com/llms.txt)\r\n\r\n## Links\r\n\r\n- **Website**: [ambermem.com](https://ambermem.com)\r\n- **MCP endpoint**: `https://mcp.ambermem.com`\r\n- **Privacy policy**: [ambermem.com/privacy](https://ambermem.com/privacy)\r\n- **Terms of service**: [ambermem.com/terms](https://ambermem.com/terms)\r\n- **Technical docs (for AI)**: [ambermem.com/llms.txt](https://ambermem.com/llms.txt)\r\n- **Report a problem**: [ambermem.com/report](https://ambermem.com/report) — no account needed, and it works even when your client cannot connect. In-session, an assistant can also call the `amber_send_feedback_to_developer` tool directly.\r\n\r\n## FAQ\r\n\r\n**Will it slow my AI down?**\r\nNo. Storage is async (background). Search adds <1 second.\r\n\r\n**What if Amber shuts down?**\r\nExport all your data as JSON anytime. Your data is always yours.\r\n\r\n**Do I need a PayPal account?**\r\nCurrently yes. PayPal handles both identity and billing. More login options coming soon.\r\n\r\n**Is my data safe?**\r\nEach user gets a completely isolated database. No data is shared between users. Amber has no access to your PayPal payment details.\r\n\r\n**Can I self-host?**\r\nNot currently. Amber is a managed service. We handle the infrastructure, scaling, and LLM costs so you don't have to.\r\n",
  "bytes": 5503,
  "sha": "cbce68dbde7a17553f2791f8e74c9ce414cb773a7454d7e81b8aa01aded35220",
  "repo_slug": "ambermem/amber",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_com_ambermem_amber_d3668c7a/readme"
}