# Agent Memory

## Docs

- [How Agent Memory Captures and Recalls Context](https://agent-memory.dev/docs/concepts/how-memory-works.md): Agent Memory hooks into every agent tool use, compresses observations into structured memories, and injects the right context at session start — automatically.
- [Memory Types and the Consolidation Pipeline](https://agent-memory.dev/docs/concepts/memory-types.md): Agent Memory organizes context into four tiers — working, episodic, semantic, and procedural — using an automated consolidation pipeline to surface what matters.
- [Hybrid Search: How Agent Memory Finds Relevant Context](https://agent-memory.dev/docs/concepts/search.md): Agent Memory uses triple-stream hybrid search combining BM25 keyword matching, vector embeddings, and knowledge graph traversal — fused with Reciprocal Rank Fusion.
- [Connect AI Coding Agents to Agent Memory in One Command](https://agent-memory.dev/docs/connect-agents.md): Wire up Claude Code, Cursor, GitHub Copilot, Codex, Gemini CLI, and 13+ other MCP-compatible agents to Agent Memory in one command.
- [Configure Agent Memory: Features, Ports, and Behavior](https://agent-memory.dev/docs/guides/configuration.md): Configure Agent Memory by editing ~/.agentmemory/.env. Enable consolidation, knowledge graphs, context injection, memory slots, and more.
- [Import Claude Code Session History into Agent Memory](https://agent-memory.dev/docs/guides/import-history.md): Bulk-import your existing Claude Code JSONL session transcripts so Agent Memory can recall context from sessions before you installed it.
- [Connect an LLM Provider to Agent Memory](https://agent-memory.dev/docs/guides/llm-providers.md): Connect Anthropic, OpenAI, Gemini, OpenRouter, or run fully offline. Agent Memory uses LLMs for memory compression, consolidation, and knowledge graph extraction.
- [Share Memory Across Your Team with Team Memory](https://agent-memory.dev/docs/guides/team-memory.md): Enable team-shared memory so multiple developers and agents share learned context, patterns, and decisions across a project — with per-agent scoping options.
- [Troubleshoot Agent Memory: Common Issues and Fixes](https://agent-memory.dev/docs/guides/troubleshooting.md): Fix common Agent Memory issues: server not starting, agent not connecting, poor recall, high token usage, and more. Use agentmemory doctor for guided diagnostics.
- [Add Persistent Memory to Claude Code](https://agent-memory.dev/docs/integrations/claude-code.md): Install Agent Memory for Claude Code using the plugin marketplace or CLI — memory persists across all sessions with automatic hook-based capture.
- [Add Persistent Memory to GitHub Copilot](https://agent-memory.dev/docs/integrations/copilot.md): Connect Agent Memory to GitHub Copilot CLI and Copilot Workspace via MCP. Copilot remembers your codebase patterns and decisions across sessions.
- [Add Persistent Memory to Cursor](https://agent-memory.dev/docs/integrations/cursor.md): Connect Agent Memory to Cursor via MCP in one command. Your Cursor Agent will remember file patterns, decisions, and context across every session.
- [Connect Any MCP-Compatible Agent to Agent Memory](https://agent-memory.dev/docs/integrations/mcp-clients.md): Add Agent Memory to Cline, Continue, Zed, Warp, OpenCode, Kiro, Droid, Qwen, Antigravity, Gemini CLI, and other MCP-compatible AI coding agents.
- [Agent Memory: Persistent Memory for AI Coding Agents](https://agent-memory.dev/docs/introduction.md): Agent Memory gives AI coding agents persistent memory across sessions with hybrid search and a 4-tier pipeline. 95.2% recall, 92% fewer tokens per session.
- [Get Started with Agent Memory in Under 2 Minutes](https://agent-memory.dev/docs/quickstart.md): Install Agent Memory, start the memory server, connect your first agent, and verify that persistent recall is working — all in under 2 minutes.
- [Memory Management API — Save, Recall, Forget, and Export](https://agent-memory.dev/docs/reference/api-memory.md): REST API endpoints for managing memories: save important context, recall past memories, forget obsolete entries, and export or import your full memory database.
- [Agent Memory REST API — Overview and Authentication](https://agent-memory.dev/docs/reference/api-overview.md): Agent Memory exposes 128 REST endpoints on port 3111. All endpoints are prefixed with /agentmemory/. Authenticate with a Bearer token when AGENTMEMORY_SECRET is set.
- [Search and Context API — Smart Search and Enrichment](https://agent-memory.dev/docs/reference/api-search.md): REST API endpoints for searching memories and generating context: hybrid smart-search, context block generation, file enrichment, and knowledge graph queries.
- [Session Management API — Start, End, and List Sessions](https://agent-memory.dev/docs/reference/api-sessions.md): REST API endpoints for managing Agent Memory sessions: start a session to get injected context, end a session to trigger consolidation, and list past sessions.
- [agentmemory CLI Commands Reference](https://agent-memory.dev/docs/reference/cli.md): Complete reference for all agentmemory CLI commands: start, init, connect, status, doctor, demo, upgrade, stop, remove, mcp, and import-jsonl.
- [Agent Memory Configuration Reference](https://agent-memory.dev/docs/reference/configuration.md): Full reference for all Agent Memory environment variables. Configure LLM providers, embedding models, feature flags, ports, search weights, and multi-agent scoping.
- [Agent Memory MCP Resources: All 6 agentmemory:// URIs](https://agent-memory.dev/docs/reference/mcp-resources.md): Reference for all 6 agentmemory:// MCP resources — read-only data streams for server status, project profiles, recent memories, and graph stats.
- [Agent Memory MCP Tools: Complete Reference (53 Tools)](https://agent-memory.dev/docs/reference/mcp-tools.md): Complete reference for all 53 Agent Memory MCP tools — core recall and save, orchestration, team sharing, knowledge graph, lessons, slots, and more.