Engram Documentation
Memory infrastructure for AI agents. Engram stores complete, uncompressed conversation transcripts and makes them instantly searchable.
Quick Links
| Getting Started | Connect in 2 minutes |
| API Reference | All 6 MCP tools with parameters and examples |
| Integrations | Claude Desktop, Cursor, Windsurf, custom clients |
Learn
| Concepts | How conversations, chunking, and search work |
| Architecture | System design, request lifecycle, and why this stack |
| Use Cases | Agent memory, support history, knowledge bases, audit trails |
| Examples | End-to-end tutorials and walkthroughs |
| Patterns | Best practices for tagging, batching, and search |
| Comparison | How Engram differs from Mem0, Zep, Supermemory |
Reference
| Authentication | API key format and security |
| Data Model | Full schema reference with field descriptions |
| Limits and Quotas | Request limits, storage limits, platform limits |
| Security | Tenant isolation, encryption, privacy |
| Glossary | Definitions of key terms |
Operations
| Troubleshooting | Common issues and fixes |
| FAQ | Frequently asked questions |
Project
| Roadmap | What’s built, what’s next |
| Changelog | Release history |
Last updated on