# Maximem Synap > Agentic Context Management ## Docs - [DOCUMENTATION GUIDELINES](https://docs.maximem.ai/DOCUMENTATION_GUIDELINES.md) - [Analytics](https://docs.maximem.ai/api-reference/analytics.md): The Analytics API provides access to usage metrics, performance data, and cost tracking for your Synap instances. Use it to monitor API usage, track latency, and understand resource consumption. - [Configuration](https://docs.maximem.ai/api-reference/configuration.md): The Memory Architecture Configuration API (MACA) controls how Synap processes, stores, and retrieves memories for an instance. Configuration changes go through a review workflow to prevent accidental misconfiguration in production. - [Context (Organizational & Customer)](https://docs.maximem.ai/api-reference/context.md): The Context API provides REST endpoints for fetching **organizational (client-scoped)** and **customer-scoped** context. These endpoints are designed for server-side operations where you need to retrieve shared knowledge that applies across users or across your entire application. - [Error Codes](https://docs.maximem.ai/api-reference/errors.md): All Synap API errors follow a consistent format with a human-readable `error` message, a machine-readable `code`, and an optional `details` object with additional context. - [Instances](https://docs.maximem.ai/api-reference/instances.md): Instances are the primary unit of isolation in Synap. Each instance represents an independent memory store for an AI agent, with its own configuration, storage, and authentication credentials. Instances belong to a client (your organization) and are managed through the Dashboard API. - [Memory (Bootstrap Ingestion)](https://docs.maximem.ai/api-reference/memory.md): The Memory API is the REST interface for ingesting documents, tracking ingestion progress, and managing stored memories. It is designed primarily for **bootstrap ingestion** — bulk loading historical data, backfilling conversations, and migrating from other systems. - [Migration API](https://docs.maximem.ai/api-reference/migration.md): API endpoints for migrating data to Synap. - [API Overview](https://docs.maximem.ai/api-reference/overview.md): The Synap REST API provides programmatic access for **bootstrap ingestion**, **organizational and customer context retrieval**, and **data migration**. All endpoints return JSON and accept JSON request bodies. - [Webhooks](https://docs.maximem.ai/api-reference/webhooks.md): Synap webhooks deliver real-time notifications to your application when events occur. Use webhooks to trigger workflows, update dashboards, or synchronize state without polling the API. - [Accurate Mode](https://docs.maximem.ai/concepts/accurate-mode.md): Accurate mode prioritizes thoroughness and quality over speed. It runs the full extraction pipeline during ingestion and combines vector search with graph traversal during retrieval, producing richer, more connected context at the cost of higher latency and compute usage. Use accurate mode when the… - [Agent Interactions](https://docs.maximem.ai/concepts/agent-interactions.md): Building a memory-enabled agent follows a consistent pattern: retrieve context, generate a response, and ingest the conversation. This cycle repeats for every interaction, gradually building a richer memory that makes each subsequent response more informed and personalized. This page walks through t… - [Bootstrap Ingestion](https://docs.maximem.ai/concepts/bootstrap-ingestion.md): Bootstrap ingestion is the process of loading pre-existing data into Synap in bulk. Before your agent starts handling live conversations, you often need to seed it with historical context: past conversations, product documentation, knowledge base articles, customer records, and other reference mater… - [Clients & Instances](https://docs.maximem.ai/concepts/clients-and-instances.md): Synap organizes memory around two fundamental primitives: **Clients** and **Instances**. A Client represents your organization, and Instances represent individual AI agent deployments within that organization. Understanding this hierarchy is essential to configuring memory isolation, authentication,… - [Context Compaction](https://docs.maximem.ai/concepts/context-compaction.md): As conversations grow longer, sending the full history to your LLM becomes expensive and eventually hits token limits. Context compaction intelligently compresses conversation history, preserving key facts, decisions, preferences, and emotional context while dramatically reducing token count. Instea… - [Conversational Context Lifecycle](https://docs.maximem.ai/concepts/conversational-context-lifecycle.md): How context accumulates, compacts, and is managed within a conversation. - [Customer Context](https://docs.maximem.ai/concepts/customer-context.md): Customer context is knowledge stored at the **CUSTOMER scope** -- shared across all users within a specific customer organization but invisible to users in other organizations. It represents the collective knowledge about a particular tenant: their policies, structure, preferences, projects, and dom… - [Customers & Users](https://docs.maximem.ai/concepts/customers-and-users.md): While [Clients and Instances](/concepts/clients-and-instances) represent the infrastructure side of Synap, **Customers** and **Users** represent the people your AI agent serves. These two entities are the primary drivers of memory scoping -- they determine what your agent remembers about whom, and w… - [Customized Memory Architectures](https://docs.maximem.ai/concepts/customized-memory-architectures.md): The Memory Architecture Configuration (MACA) is a YAML-based configuration system that controls every aspect of how a Synap Instance handles memory. It defines where memories are stored, how incoming data is processed, and how memories are retrieved. Each Instance has its own MACA with independent v… - [Entity Resolution & Master Data Management](https://docs.maximem.ai/concepts/entity-resolution.md): Entity Resolution (ER) is Synap's ability to identify and link mentions of the same real-world entity across different conversations and documents. When a user says "John", "Mr. Smith", "my manager", and "the person I met at the conference", Synap determines whether these all refer to the same indiv… - [Fast Mode](https://docs.maximem.ai/concepts/fast-mode.md): Fast mode prioritizes speed over thoroughness for both ingestion and retrieval. It is the recommended default for real-time conversational agents where low latency matters more than exhaustive extraction or relationship-aware context. Most production agents use fast mode for the majority of their in… - [Integration Overview](https://docs.maximem.ai/concepts/integration-overview.md): A map of all Synap integration points and how they connect. - [Long-term Context](https://docs.maximem.ai/concepts/long-term-context.md): Long-term context is the persistent knowledge layer in Synap. Unlike [short-term context](/concepts/short-term-context), which lives only for the duration of a conversation, long-term context persists across sessions -- days, weeks, months, and years. It is the accumulated knowledge that your AI age… - [Long-term Context Lifecycle](https://docs.maximem.ai/concepts/long-term-context-lifecycle.md): How memories are extracted, stored, aged, and retrieved across conversations. - [Memories & Context](https://docs.maximem.ai/concepts/memories-and-context.md): At the heart of Synap are two complementary concepts: **Memories** and **Context**. Memories are the structured knowledge that Synap extracts and stores from your data. Context is the curated set of memories, assembled and delivered to your AI agent at the moment it needs them. Understanding the rel… - [Memory Scopes](https://docs.maximem.ai/concepts/memory-scopes.md): Scopes define memory isolation boundaries in Synap. They determine who can see which memories and how memories are organized across your application's user hierarchy. Getting scopes right is critical for building multi-user, multi-tenant AI applications where personal data stays personal and shared… - [Memory Types](https://docs.maximem.ai/concepts/memory-types.md): Synap's extraction pipeline does not store raw text. Instead, it analyzes every ingested document and produces **structured, typed memories** -- discrete units of knowledge that capture different dimensions of meaning. There are five memory types, each designed to represent a specific kind of inform… - [Org-context Lifecycle](https://docs.maximem.ai/concepts/org-context-lifecycle.md): How organizational knowledge enters, is stored, and surfaces in retrieval. - [Organizational Context](https://docs.maximem.ai/concepts/organizational-context.md): Organizational context is knowledge stored at the **CLIENT scope** -- the broadest application-level scope in Synap's hierarchy. It represents your product's documentation, feature announcements, domain knowledge, and any other information that should be accessible to every user across every custome… - [Runtime Ingestion](https://docs.maximem.ai/concepts/runtime-ingestion.md): Runtime ingestion is the process of feeding data into Synap as it is generated during live agent interactions. This is the primary ingestion path for most applications. After each conversation turn -- or at the end of a conversation -- your application calls the SDK to send the content through Synap… - [Short-term Context](https://docs.maximem.ai/concepts/short-term-context.md): Short-term context is the accumulated conversation history within a single session. It is the memory of what has been said in the current conversation -- the questions asked, the answers given, the decisions made, and the topics discussed. Short-term context is what allows your AI agent to maintain… - [Connectors](https://docs.maximem.ai/connectors/overview.md): Pre-built integrations for pulling data from external systems into Synap. - [Managing Instances](https://docs.maximem.ai/dashboard/managing-instances.md): Instances are the fundamental deployment unit in Synap. Each instance is an isolated memory agent with its own storage namespaces, memory architecture configuration, scope hierarchy, and API keys. You can think of an instance as a dedicated memory environment for a specific agent or use case within… - [Memory Configuration](https://docs.maximem.ai/dashboard/memory-configuration.md): Memory Architecture Configuration (MACA) is the central configuration system that controls how each Synap instance stores, ingests, and retrieves memories. MACA is defined in YAML, managed through the dashboard with full version control, and enforced through an approval workflow before activation. - [Monitoring & Analytics](https://docs.maximem.ai/dashboard/monitoring-and-analytics.md): Track ingestion throughput, context retrieval usage, and system health. - [Dashboard Overview](https://docs.maximem.ai/dashboard/overview.md): The Synap Dashboard is a web-based management interface for your Synap deployment. From a single interface at [synap.maximem.ai](https://synap.maximem.ai), you can create and manage instances and configure memory architecture. - [Team Management](https://docs.maximem.ai/dashboard/team-management.md): Manage team members and access to your Synap workspace. - [Webhooks](https://docs.maximem.ai/dashboard/webhooks.md): Receive real-time event notifications from Synap. - [What is Synap?](https://docs.maximem.ai/getting-started/overview.md): Synap is a managed memory layer for AI agents. It sits between your application and your LLM, providing persistent structured memory that survives across sessions, conversations, and deployments. Instead of treating every conversation as a blank slate, your agents can remember, learn, and personaliz… - [Quickstart](https://docs.maximem.ai/getting-started/quickstart.md): This guide walks you through installing the Synap SDK, creating your first instance, ingesting a memory, and retrieving context. By the end, you'll have a working memory pipeline in under 10 minutes. - [Configuring Memory Architecture](https://docs.maximem.ai/guides/configuring-memory.md): The **Memory Architecture Configuration Artifact** (MACA) is a YAML document that controls every aspect of how Synap processes, stores, and retrieves memories for an instance. It determines what gets extracted from conversations, how it is organized in storage, and how retrieval results are ranked a… - [Build Your First Integration](https://docs.maximem.ai/guides/first-integration.md): This tutorial walks you through building a complete memory-enabled chatbot from scratch. By the end, you will have a FastAPI application that remembers user preferences, facts, and context across conversations -- powered by Synap's persistent memory layer. - [Migration Guide](https://docs.maximem.ai/guides/migration.md): This guide covers the full spectrum of Synap migration scenarios: upgrading the SDK, migrating MACA configurations, changing scope strategies, switching embedding models, and moving from development to production. Each section includes step-by-step procedures and rollback strategies. - [Multi-User Memory Scoping](https://docs.maximem.ai/guides/multi-user-scoping.md): Most real-world applications serve multiple users, often across multiple organizations. Synap's scoping system ensures that memories are properly isolated while still enabling shared context where appropriate. This guide explains the scope hierarchy, walks through common patterns, and shows you how… - [Production Checklist](https://docs.maximem.ai/guides/production-checklist.md): This checklist covers every aspect of a production-ready Synap integration -- from security and SDK configuration to monitoring and operational procedures. Work through each section before your first production deployment, and revisit it before subsequent releases. - [Use-Case Markdown](https://docs.maximem.ai/guides/use-case-markdown.md): The Use-Case Markdown file tells Synap what your agent does, who it serves, and what it should remember. Synap uses it to generate and continuously refine the Memory Architecture Configuration (MACA) for your instance. - [Synap Developer Documentation](https://docs.maximem.ai/index.md): Synap gives your AI agents long-term memory. Ingest conversations, extract structured knowledge, and retrieve contextual memories — all through a simple SDK. No infrastructure to manage, no vector databases to tune, no retrieval pipelines to build. - [Agno](https://docs.maximem.ai/integrations/agno.md): Drop-in InMemoryDb replacement that routes Agno user memories through Synap. - [AutoGen](https://docs.maximem.ai/integrations/autogen.md): BaseTool implementations for memory search and storage in AutoGen agents. - [Claude Agent SDK](https://docs.maximem.ai/integrations/claude-agent.md): Hooks and MCP server for Anthropic's Claude Agent SDK — available in Python and TypeScript. - [CrewAI](https://docs.maximem.ai/integrations/crewai.md): StorageBackend implementation that routes CrewAI memory through Synap. - [Google ADK](https://docs.maximem.ai/integrations/google-adk.md): FunctionTool factory for Google Agent Development Kit agents. - [Haystack](https://docs.maximem.ai/integrations/haystack.md): SynapRetriever and SynapMemoryWriter pipeline components for Haystack. - [LangChain](https://docs.maximem.ai/integrations/langchain.md): Memory, callbacks, retriever, and tools for LangChain chains and agents. - [LangGraph](https://docs.maximem.ai/integrations/langgraph.md): Checkpointer and cross-thread store for LangGraph graphs powered by Synap memory. - [LiveKit Agents](https://docs.maximem.ai/integrations/livekit-agents.md): Memory preloading and turn recording for LiveKit voice agents. - [LlamaIndex](https://docs.maximem.ai/integrations/llamaindex.md): BaseMemory implementation and semantic retriever for LlamaIndex pipelines. - [Mastra](https://docs.maximem.ai/integrations/mastra.md): SynapMemory class and tools for the Mastra ADK — TypeScript. - [Microsoft Agent Framework](https://docs.maximem.ai/integrations/microsoft-agent.md): Context and history providers for the Microsoft Agent Framework (MAF). - [NeMo Agent Toolkit](https://docs.maximem.ai/integrations/nemo-agent-toolkit.md): MemoryEditor implementation for NVIDIA NeMo Agent Toolkit (NAT) workflows. - [OpenAI Agents SDK](https://docs.maximem.ai/integrations/openai-agents.md): Search and store function tools for the OpenAI Agents SDK. - [Integrations Overview](https://docs.maximem.ai/integrations/overview.md): Drop-in packages that add Synap memory to popular AI frameworks and agent SDKs. - [Pipecat](https://docs.maximem.ai/integrations/pipecat.md): Frame processors for memory injection and turn recording in Pipecat voice pipelines. - [Pydantic AI](https://docs.maximem.ai/integrations/pydantic-ai.md): Dependency dataclass and auto-registered tools for Pydantic AI agents. - [Semantic Kernel](https://docs.maximem.ai/integrations/semantic-kernel.md): Kernel plugin with search and store functions for Microsoft Semantic Kernel. - [Vercel AI SDK](https://docs.maximem.ai/integrations/vercel-adk.md): Model middleware that wraps any Vercel AI SDK model with automatic Synap context. - [Migration](https://docs.maximem.ai/migration/overview.md): This reference covers the full spectrum of Synap migration scenarios: upgrading the SDK, migrating MACA configurations, changing scope strategies, switching embedding models, moving from development to production, importing data from other memory systems, and performing bulk data imports. Each secti… - [OpenClaw](https://docs.maximem.ai/plugins/openclaw.md): Memory plugin for OpenClaw — syncs AI context across OpenClaw, ChatGPT, Claude, Gemini, Manus, and more. - [Plugins Overview](https://docs.maximem.ai/plugins/overview.md): First-party plugins that bring Maximem Synap memory into third-party agents and tools. - [Changelog](https://docs.maximem.ai/resources/changelog.md): All notable changes to the Synap SDK and API are documented here. This project follows [Semantic Versioning](https://semver.org/) (SemVer). - [FAQ](https://docs.maximem.ai/resources/faq.md): Common questions about Synap, organized by topic. If your question is not answered here, check the [Support](/resources/support) page for additional help channels. - [Glossary](https://docs.maximem.ai/resources/glossary.md): A comprehensive reference of terms, concepts, and identifiers used throughout the Synap platform. - [Support](https://docs.maximem.ai/resources/support.md): Whether you are stuck on an integration, have a question about the API, or need to report an issue, we are here to help. - [SDK Configuration](https://docs.maximem.ai/sdk/configuration.md): Customize SDK behavior for your environment. - [Context Compaction](https://docs.maximem.ai/sdk/context-compaction.md): Compress conversations while preserving key information. - [Context Fetch](https://docs.maximem.ai/sdk/context-fetch.md): Fetch contextual memories for your AI agent. Also known as memory retrieval. - [Entity Resolution](https://docs.maximem.ai/sdk/entity-resolution.md): Working with entity resolution in the SDK. - [Error Handling](https://docs.maximem.ai/sdk/error-handling.md): Handle errors gracefully in your Synap integration. - [Ingestion](https://docs.maximem.ai/sdk/ingestion.md): Send conversations, documents, and other data into Synap. - [Initializing the SDK](https://docs.maximem.ai/sdk/initialization.md): Set up the Synap SDK in your application. - [Authentication](https://docs.maximem.ai/setup/authentication.md): How to authenticate your SDK with Synap Cloud using API keys. - [Installation](https://docs.maximem.ai/setup/installation.md): This page covers installing the Synap SDK, configuring environment variables, and verifying your setup. - [Integration](https://docs.maximem.ai/setup/integration.md): Connect Synap to your application framework and LLM provider. - [Storage Infrastructure](https://docs.maximem.ai/setup/storage-infrastructure.md): Synap uses two complementary storage engines to persist and retrieve memories: a **vector store** for semantic similarity search and a **graph store** for relationship-based queries. Both engines are configured through the Memory Architecture Configuration (MACA) and work together to provide compreh… ## Optional - [GitHub](https://github.com/synap-dev/synap-sdk) - [Community](https://discord.gg/synap)