> ## Documentation Index
> Fetch the complete documentation index at: https://agent-memory.dev/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Agent Memory: Persistent Memory for AI Coding Agents

> Agent Memory gives your AI coding agents persistent memory across sessions — with 95.2% retrieval recall and ~92% token reduction compared to full-context approaches.

Agent Memory is a persistent memory engine for AI coding agents. It silently runs in the background, capturing everything your agent observes — file edits, commands, decisions, discoveries — and recalls the right context at the start of every new session. Your agents stop forgetting.

<CardGroup cols={2}>
  <Card title="Quick Start" icon="rocket" href="/docs/quickstart">
    Install Agent Memory and connect your first agent in under 2 minutes.
  </Card>

  <Card title="Connect Agents" icon="plug" href="/docs/connect-agents">
    Wire up Claude Code, Cursor, Copilot, Codex, and 13+ other agents.
  </Card>

  <Card title="Core Concepts" icon="brain" href="/docs/concepts/how-memory-works">
    Understand how Agent Memory captures, compresses, and recalls context.
  </Card>

  <Card title="Configuration" icon="sliders" href="/docs/guides/configuration">
    Tune memory behavior, enable features, and connect LLM providers.
  </Card>
</CardGroup>

## Get up and running

<Steps>
  <Step title="Install Agent Memory">
    ```bash theme={null}
    npm install -g @agentmemory/agentmemory
    ```
  </Step>

  <Step title="Start the memory server">
    ```bash theme={null}
    agentmemory
    ```

    The server starts on port 3111 and a real-time viewer launches on port 3113.
  </Step>

  <Step title="Connect your agent">
    For Claude Code, use the plugin marketplace:

    ```
    /plugin marketplace add rohitg00/agentmemory
    /plugin install agentmemory
    ```

    For all other MCP-compatible agents, run:

    ```bash theme={null}
    agentmemory connect cursor   # or copilot, codex, gemini-cli, cline, etc.
    ```
  </Step>

  <Step title="Verify it's working">
    ```bash theme={null}
    agentmemory status
    ```

    You'll see session counts, memory stats, and token savings in real time.
  </Step>
</Steps>

## Why Agent Memory

<CardGroup cols={3}>
  <Card title="95.2% Recall" icon="bullseye">
    Best-in-class retrieval on LongMemEval-S (500 questions), using triple-stream hybrid search: BM25 + vector + knowledge graph.
  </Card>

  <Card title="92% Token Savings" icon="bolt">
    \~170K tokens per year vs 19.5M+ for full-context approaches — without losing relevant history.
  </Card>

  <Card title="No External DBs" icon="database">
    Runs entirely on local SQLite. No Redis, no Postgres, no cloud setup required.
  </Card>
</CardGroup>

## Explore the docs

<CardGroup cols={2}>
  <Card title="How Memory Works" icon="diagram-project" href="/docs/concepts/how-memory-works">
    Learn about the 4-tier consolidation pipeline and how observations become memories.
  </Card>

  <Card title="LLM Providers" icon="sparkles" href="/docs/guides/llm-providers">
    Connect Anthropic, OpenAI, Gemini, OpenRouter, or run fully offline with Ollama.
  </Card>

  <Card title="Team Memory" icon="users" href="/docs/guides/team-memory">
    Share memory across your team with multi-agent scoping and coordination.
  </Card>

  <Card title="REST API" icon="code" href="/docs/reference/api-overview">
    Integrate directly with the 128-endpoint REST API on port 3111.
  </Card>

  <Card title="MCP Tools" icon="wrench" href="/docs/reference/mcp-tools">
    Reference for all 53 MCP tools, resources, and prompts.
  </Card>

  <Card title="CLI Reference" icon="terminal" href="/docs/reference/cli">
    Full reference for all agentmemory CLI commands and options.
  </Card>
</CardGroup>