--- title: "#  Core concepts" section: "raw" type: "source" created: "2026-08-30" updated: "2026-08-30" canonical: "https://pyweb.dev/wiki/raw/articles/anthropic-claude-managed-agents-overview-2026" --- Copy page  Anthropic offers two ways to build with Claude, each suited to different use cases: | | Messages API | Claude Managed Agents | | --- | --- | --- | | **What it is** | Direct model prompting access | Pre-built, configurable agent harness that runs in managed infrastructure | | **Best for** | Custom agent loops and fine-grained control | Long-running tasks and asynchronous work | Claude Managed Agents provides the harness and infrastructure for running Claude as an autonomous agent. Instead of building your own agent loop, tool execution, and runtime, you get a fully managed environment where Claude can read files, run commands, browse the web, and run code securely. The harness supports built-in prompt caching, compaction, and other performance optimizations for high-quality, efficient agent outputs. To build your own agent loop with direct model access instead, see [Using the Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages).  [Quickstart](https://platform.claude.com/docs/en/managed-agents/quickstart) Create your first agent session  [Start a session](https://platform.claude.com/docs/en/managed-agents/sessions) Create a session and send your first event  [Reference](https://platform.claude.com/docs/en/managed-agents/reference) Event types, rate limits, CLI flags, and other lookup tables ##  Core concepts Claude Managed Agents is built around four concepts: | Concept | Description | | --- | --- | | **Agent** | The model, system prompt, tools, MCP servers, and skills | | **Environment** | Configuration for where sessions run: an Anthropic-managed cloud sandbox, or a self-hosted sandbox on your own infrastructure | | **Session** | A running agent instance within an environment, performing a specific task and generating outputs | | **Events** | Messages exchanged between your application and the agent (user turns, tool results, status updates) | ##  How it works 1. 1 Create an agent Define the model, system prompt, tools, MCP servers, and skills. Create the agent once and reference it by ID across sessions. 2. 2 Create an environment Configure where the agent runs: a cloud sandbox, or a [self-hosted sandbox](https://platform.claude.com/docs/en/managed-agents/self-hosted-sandboxes) on your own infrastructure. 3. 3 Start a session Launch a session that references your agent and environment configuration. 4. 4 Send events and stream responses Send user messages as events. Claude autonomously runs tools and streams back results through server-sent events (SSE). Event history is persisted server-side and can be fetched in full. 5. 5 Steer or interrupt Send additional user events to guide the agent mid-execution, or interrupt it to change direction. ##  When to use Claude Managed Agents Claude Managed Agents is best for workloads that need: - **Long-running execution:** Tasks that run for minutes or hours with multiple tool calls - **Cloud infrastructure:** Secure sandboxes with pre-installed packages and network access - **Self-hosted execution:** Sandboxes on infrastructure you control for compliance or data-residency requirements - **Minimal infrastructure:** No need to build your own agent loop, sandbox, or tool execution layer - **Stateful sessions:** Persistent filesystems and conversation history across multiple interactions - **Scheduled execution:** Recurring agent runs on a cron schedule through [scheduled deployments](https://platform.claude.com/docs/en/managed-agents/scheduled-deployments) ##  Supported tools Claude Managed Agents gives Claude access to a set of built-in tools: - **Bash:** Run shell commands in the sandbox - **File operations:** Read, write, edit, glob, and grep files in the sandbox - **Web search and fetch:** Search the web and retrieve content from URLs, optionally restricted to an allowlist or blocklist of domains - **MCP servers:** Connect to external tool providers See [Tools](https://platform.claude.com/docs/en/managed-agents/tools) for the full list and configuration options. ##  Beta access To get started, you need: 1. A [Claude API key](https://platform.claude.com/settings/keys) 2. The `managed-agents-2026-04-01` beta header on all requests 3. Access to Claude Managed Agents (enabled by default for all API accounts) Within the beta, [MCP tunnels](https://platform.claude.com/docs/en/agents-and-tools/mcp-tunnels/overview) and [dreaming](https://platform.claude.com/docs/en/managed-agents/dreams) are in a more limited research preview. [Request access](https://claude.com/form/claude-managed-agents) to enable them. Claude Managed Agents is stateful by design: sessions are long-running, resume cleanly after pauses, and store conversation history, sandbox state, and outputs server-side. Because of this, Managed Agents is not currently eligible for [Zero Data Retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#zero-data-retention-zdr-scope) or HIPAA Business Associate Agreement (BAA) coverage. You retain control over this data: you can [delete sessions](https://platform.claude.com/docs/en/managed-agents/session-operations#deleting-a-session), and separately delete any [files](https://platform.claude.com/docs/en/build-with-claude/files#delete-a-file) you uploaded, at any time through the API. For eligibility across all features, see [API and data retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#feature-eligibility). See [Rate limits](https://platform.claude.com/docs/en/managed-agents/reference#rate-limits) and [Branding guidelines](https://platform.claude.com/docs/en/managed-agents/reference#branding-guidelines) in the reference. Was this page helpful?  Claude Managed Agents overview/ Core concepts Ask Docs --- ## Agent Navigation ### Machine endpoints - Knowledge graph: https://pyweb.dev/api/graph.json - Graph analysis: https://pyweb.dev/api/graph-analysis.json - Context index: https://pyweb.dev/llms.txt