## Create Agent `BetaManagedAgentsAgent Beta.Agents.Create(AgentCreateParamsparameters, CancellationTokencancellationToken = default)` **post** `/v1/agents` Create Agent ### Parameters - `AgentCreateParams parameters` - `required Model model` Body param: Model identifier. Accepts the [model string](https://platform.claude.com/docs/en/about-claude/models/overview#latest-models-comparison), e.g. `claude-opus-4-6`, or a `model_config` object for additional configuration control - `enum BetaManagedAgentsModel:` The model that will power your agent. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options. - `"claude-opus-4-8"ClaudeOpus4_8` Frontier intelligence for long-running agents and coding - `"claude-opus-4-7"ClaudeOpus4_7` Frontier intelligence for long-running agents and coding - `"claude-opus-4-6"ClaudeOpus4_6` Most intelligent model for building agents and coding - `"claude-sonnet-4-6"ClaudeSonnet4_6` Best combination of speed and intelligence - `"claude-haiku-4-5"ClaudeHaiku4_5` Fastest model with near-frontier intelligence - `"claude-haiku-4-5-20251001"ClaudeHaiku4_5_20251001` Fastest model with near-frontier intelligence - `"claude-opus-4-5"ClaudeOpus4_5` Premium model combining maximum intelligence with practical performance - `"claude-opus-4-5-20251101"ClaudeOpus4_5_20251101` Premium model combining maximum intelligence with practical performance - `"claude-sonnet-4-5"ClaudeSonnet4_5` High-performance model for agents and coding - `"claude-sonnet-4-5-20250929"ClaudeSonnet4_5_20250929` High-performance model for agents and coding - `class BetaManagedAgentsModelConfigParams:` An object that defines additional configuration control over model use - `required BetaManagedAgentsModel ID` The model that will power your agent. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options. - `"claude-opus-4-8"ClaudeOpus4_8` Frontier intelligence for long-running agents and coding - `"claude-opus-4-7"ClaudeOpus4_7` Frontier intelligence for long-running agents and coding - `"claude-opus-4-6"ClaudeOpus4_6` Most intelligent model for building agents and coding - `"claude-sonnet-4-6"ClaudeSonnet4_6` Best combination of speed and intelligence - `"claude-haiku-4-5"ClaudeHaiku4_5` Fastest model with near-frontier intelligence - `"claude-haiku-4-5-20251001"ClaudeHaiku4_5_20251001` Fastest model with near-frontier intelligence - `"claude-opus-4-5"ClaudeOpus4_5` Premium model combining maximum intelligence with practical performance - `"claude-opus-4-5-20251101"ClaudeOpus4_5_20251101` Premium model combining maximum intelligence with practical performance - `"claude-sonnet-4-5"ClaudeSonnet4_5` High-performance model for agents and coding - `"claude-sonnet-4-5-20250929"ClaudeSonnet4_5_20250929` High-performance model for agents and coding - `Speed? Speed` Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time. - `"standard"Standard` - `"fast"Fast` - `required string name` Body param: Human-readable name for the agent. 1-256 characters. - `string? description` Body param: Description of what the agent does. Up to 2048 characters. - `IReadOnlyList mcpServers` Body param: MCP servers this agent connects to. Maximum 20. Names must be unique within the array. - `required string Name` Unique name for this server, referenced by mcp_toolset configurations. 1-255 characters. - `required Type Type` - `"url"Url` - `required string Url` Endpoint URL for the MCP server. - `IReadOnlyDictionary metadata` Body param: Arbitrary key-value metadata. Maximum 16 pairs, keys up to 64 chars, values up to 512 chars. - `BetaManagedAgentsMultiagentParams? multiagent` Body param: A coordinator topology: the session's primary thread orchestrates work by spawning session threads, each running an agent drawn from the `agents` roster. - `IReadOnlyList skills` Body param: Skills available to the agent. Maximum 20. - `class BetaManagedAgentsAnthropicSkillParams:` An Anthropic-managed skill. - `required string SkillID` Identifier of the Anthropic skill (e.g., "xlsx"). - `required Type Type` - `"anthropic"Anthropic` - `string? Version` Version to pin. Defaults to latest if omitted. - `class BetaManagedAgentsCustomSkillParams:` A user-created custom skill. - `required string SkillID` Tagged ID of the custom skill (e.g., "skill_01XJ5..."). - `required Type Type` - `"custom"Custom` - `string? Version` Version to pin. Defaults to latest if omitted. - `string? system` Body param: System prompt for the agent. Up to 100,000 characters. - `IReadOnlyList tools` Body param: Tool configurations available to the agent. Maximum of 128 tools across all toolsets allowed. - `class BetaManagedAgentsAgentToolset20260401Params:` Configuration for built-in agent tools. Use this to enable or disable groups of tools available to the agent. - `required Type Type` - `"agent_toolset_20260401"AgentToolset20260401` - `IReadOnlyList Configs` Per-tool configuration overrides. - `required Name Name` Built-in agent tool identifier. - `"bash"Bash` - `"edit"Edit` - `"read"Read` - `"write"Write` - `"glob"Glob` - `"grep"Grep` - `"web_fetch"WebFetch` - `"web_search"WebSearch` - `Boolean? Enabled` Whether this tool is enabled and available to Claude. Overrides the default_config setting. - `PermissionPolicy? PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `required Type Type` - `"always_allow"AlwaysAllow` - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `required Type Type` - `"always_ask"AlwaysAsk` - `BetaManagedAgentsAgentToolsetDefaultConfigParams? DefaultConfig` Default configuration for all tools in a toolset. - `Boolean? Enabled` Whether tools are enabled and available to Claude by default. Defaults to true if not specified. - `PermissionPolicy? PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `class BetaManagedAgentsMcpToolsetParams:` Configuration for tools from an MCP server defined in `mcp_servers`. - `required string McpServerName` Name of the MCP server. Must match a server name from the mcp_servers array. 1-255 characters. - `required Type Type` - `"mcp_toolset"McpToolset` - `IReadOnlyList Configs` Per-tool configuration overrides. - `required string Name` Name of the MCP tool to configure. 1-128 characters. - `Boolean? Enabled` Whether this tool is enabled. Overrides the `default_config` setting. - `PermissionPolicy? PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `BetaManagedAgentsMcpToolsetDefaultConfigParams? DefaultConfig` Default configuration for all tools from an MCP server. - `Boolean? Enabled` Whether tools are enabled by default. Defaults to true if not specified. - `PermissionPolicy? PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `class BetaManagedAgentsCustomToolParams:` A custom tool that is executed by the API client rather than the agent. When the agent calls this tool, an `agent.custom_tool_use` event is emitted and the session goes idle, waiting for the client to provide the result via a `user.custom_tool_result` event. - `required string Description` Description of what the tool does, shown to the agent to help it decide when to use the tool. 1-1024 characters. - `required BetaManagedAgentsCustomToolInputSchema InputSchema` JSON Schema for custom tool input parameters. - `IReadOnlyDictionary? Properties` JSON Schema properties defining the tool's input parameters. - `IReadOnlyList Required` List of required property names. - `Type Type` Must be 'object' for tool input schemas. - `"object"Object` - `required string Name` Unique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens. - `required Type Type` - `"custom"Custom` - `IReadOnlyList betas` Header param: Optional header to specify the beta version(s) you want to use. - `"message-batches-2024-09-24"MessageBatches2024_09_24` - `"prompt-caching-2024-07-31"PromptCaching2024_07_31` - `"computer-use-2024-10-22"ComputerUse2024_10_22` - `"computer-use-2025-01-24"ComputerUse2025_01_24` - `"pdfs-2024-09-25"Pdfs2024_09_25` - `"token-counting-2024-11-01"TokenCounting2024_11_01` - `"token-efficient-tools-2025-02-19"TokenEfficientTools2025_02_19` - `"output-128k-2025-02-19"Output128k2025_02_19` - `"files-api-2025-04-14"FilesApi2025_04_14` - `"mcp-client-2025-04-04"McpClient2025_04_04` - `"mcp-client-2025-11-20"McpClient2025_11_20` - `"dev-full-thinking-2025-05-14"DevFullThinking2025_05_14` - `"interleaved-thinking-2025-05-14"InterleavedThinking2025_05_14` - `"code-execution-2025-05-22"CodeExecution2025_05_22` - `"extended-cache-ttl-2025-04-11"ExtendedCacheTtl2025_04_11` - `"context-1m-2025-08-07"Context1m2025_08_07` - `"context-management-2025-06-27"ContextManagement2025_06_27` - `"model-context-window-exceeded-2025-08-26"ModelContextWindowExceeded2025_08_26` - `"skills-2025-10-02"Skills2025_10_02` - `"fast-mode-2026-02-01"FastMode2026_02_01` - `"output-300k-2026-03-24"Output300k2026_03_24` - `"user-profiles-2026-03-24"UserProfiles2026_03_24` - `"advisor-tool-2026-03-01"AdvisorTool2026_03_01` - `"managed-agents-2026-04-01"ManagedAgents2026_04_01` - `"cache-diagnosis-2026-04-07"CacheDiagnosis2026_04_07` - `"thinking-token-count-2026-05-13"ThinkingTokenCount2026_05_13` ### Returns - `class BetaManagedAgentsAgent:` A Managed Agents `agent`. - `required string ID` - `required DateTimeOffset? ArchivedAt` A timestamp in RFC 3339 format - `required DateTimeOffset CreatedAt` A timestamp in RFC 3339 format - `required string? Description` - `required IReadOnlyList McpServers` - `required string Name` - `required Type Type` - `"url"Url` - `required string Url` - `required IReadOnlyDictionary Metadata` - `required BetaManagedAgentsModelConfig Model` Model identifier and configuration. - `required BetaManagedAgentsModel ID` The model that will power your agent. See [models](https://docs.anthropic.com/en/docs/models-overview) for additional details and options. - `"claude-opus-4-8"ClaudeOpus4_8` Frontier intelligence for long-running agents and coding - `"claude-opus-4-7"ClaudeOpus4_7` Frontier intelligence for long-running agents and coding - `"claude-opus-4-6"ClaudeOpus4_6` Most intelligent model for building agents and coding - `"claude-sonnet-4-6"ClaudeSonnet4_6` Best combination of speed and intelligence - `"claude-haiku-4-5"ClaudeHaiku4_5` Fastest model with near-frontier intelligence - `"claude-haiku-4-5-20251001"ClaudeHaiku4_5_20251001` Fastest model with near-frontier intelligence - `"claude-opus-4-5"ClaudeOpus4_5` Premium model combining maximum intelligence with practical performance - `"claude-opus-4-5-20251101"ClaudeOpus4_5_20251101` Premium model combining maximum intelligence with practical performance - `"claude-sonnet-4-5"ClaudeSonnet4_5` High-performance model for agents and coding - `"claude-sonnet-4-5-20250929"ClaudeSonnet4_5_20250929` High-performance model for agents and coding - `Speed Speed` Inference speed mode. `fast` provides significantly faster output token generation at premium pricing. Not all models support `fast`; invalid combinations are rejected at create time. - `"standard"Standard` - `"fast"Fast` - `required BetaManagedAgentsMultiagent? Multiagent` Resolved coordinator topology with a concrete agent roster. - `required IReadOnlyList Agents` Agents the coordinator may spawn as session threads, each resolved to a specific version. - `required string ID` - `required Type Type` - `"agent"Agent` - `required Int Version` - `required Type Type` - `"coordinator"Coordinator` - `required string Name` - `required IReadOnlyList Skills` - `class BetaManagedAgentsAnthropicSkill:` A resolved Anthropic-managed skill. - `required string SkillID` - `required Type Type` - `"anthropic"Anthropic` - `required string Version` - `class BetaManagedAgentsCustomSkill:` A resolved user-created custom skill. - `required string SkillID` - `required Type Type` - `"custom"Custom` - `required string Version` - `required string? System` - `required IReadOnlyList Tools` - `class BetaManagedAgentsAgentToolset20260401:` - `required IReadOnlyList Configs` - `required Boolean Enabled` - `required Name Name` Built-in agent tool identifier. - `"bash"Bash` - `"edit"Edit` - `"read"Read` - `"write"Write` - `"glob"Glob` - `"grep"Grep` - `"web_fetch"WebFetch` - `"web_search"WebSearch` - `required PermissionPolicy PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `required Type Type` - `"always_allow"AlwaysAllow` - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `required Type Type` - `"always_ask"AlwaysAsk` - `required BetaManagedAgentsAgentToolsetDefaultConfig DefaultConfig` Resolved default configuration for agent tools. - `required Boolean Enabled` - `required PermissionPolicy PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `required Type Type` - `"agent_toolset_20260401"AgentToolset20260401` - `class BetaManagedAgentsMcpToolset:` - `required IReadOnlyList Configs` - `required Boolean Enabled` - `required string Name` - `required PermissionPolicy PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `required BetaManagedAgentsMcpToolsetDefaultConfig DefaultConfig` Resolved default configuration for all tools from an MCP server. - `required Boolean Enabled` - `required PermissionPolicy PermissionPolicy` Permission policy for tool execution. - `class BetaManagedAgentsAlwaysAllowPolicy:` Tool calls are automatically approved without user confirmation. - `class BetaManagedAgentsAlwaysAskPolicy:` Tool calls require user confirmation before execution. - `required string McpServerName` - `required Type Type` - `"mcp_toolset"McpToolset` - `class BetaManagedAgentsCustomTool:` A custom tool as returned in API responses. - `required string Description` - `required BetaManagedAgentsCustomToolInputSchema InputSchema` JSON Schema for custom tool input parameters. - `IReadOnlyDictionary? Properties` JSON Schema properties defining the tool's input parameters. - `IReadOnlyList Required` List of required property names. - `Type Type` Must be 'object' for tool input schemas. - `"object"Object` - `required string Name` - `required Type Type` - `"custom"Custom` - `required Type Type` - `"agent"Agent` - `required DateTimeOffset UpdatedAt` A timestamp in RFC 3339 format - `required Int Version` The agent's current version. Starts at 1 and increments when the agent is modified. ### Example ```csharp AgentCreateParams parameters = new() { Model = BetaManagedAgentsModel.ClaudeSonnet4_6, Name = "My First Agent", }; var betaManagedAgentsAgent = await client.Beta.Agents.Create(parameters); Console.WriteLine(betaManagedAgentsAgent); ``` #### Response ```json { "id": "agent_011CZkYpogX7uDKUyvBTophP", "archived_at": null, "created_at": "2026-03-15T10:00:00Z", "description": "A general-purpose starter agent.", "mcp_servers": [ { "name": "example-mcp", "type": "url", "url": "https://example-server.modelcontextprotocol.io/sse" } ], "metadata": { "foo": "bar" }, "model": { "id": "claude-sonnet-4-6", "speed": "standard" }, "multiagent": { "agents": [ { "id": "agent_011CZkYqphY8vELVzwCUpqiQ", "type": "agent", "version": 1 } ], "type": "coordinator" }, "name": "My First Agent", "skills": [ { "skill_id": "xlsx", "type": "anthropic", "version": "1" }, { "skill_id": "skill_011CZkZFNu9hAbo3jZPRgTlx", "type": "custom", "version": "2" } ], "system": "You are a general-purpose agent that can research, write code, run commands, and use connected tools to complete the user's task end to end.", "tools": [ { "configs": [ { "enabled": true, "name": "bash", "permission_policy": { "type": "always_allow" } } ], "default_config": { "enabled": true, "permission_policy": { "type": "always_ask" } }, "type": "agent_toolset_20260401" } ], "type": "agent", "updated_at": "2026-03-15T10:00:00Z", "version": 1 } ```