Agents
Get Agent
beta.agents.retrieve(agent_id, **kwargs) -> BetaManagedAgentsAgentUpdate Agent
beta.agents.update(agent_id, **kwargs) -> BetaManagedAgentsAgentArchive Agent
beta.agents.archive(agent_id, **kwargs) -> BetaManagedAgentsAgentModels
class BetaManagedAgentsAdvisor: type: "advisor"; modelPlatform advisor roster entry: a model the session's primary thread may consult mid-turn.
Platform advisor roster entry: a model the session's primary thread may consult mid-turn.
The advisor model id.
class BetaManagedAgentsAgent: type: "agent"; id; archived_at; ... # 12 moreA Managed Agents agent.
A Managed Agents agent.
class BetaManagedAgentsAgentReference: type: "agent"; id; versionA resolved agent reference with a concrete version.
A resolved agent reference with a concrete version.
version: int
type BetaManagedAgentsAgentToolConfig = ...Configuration for a specific agent tool.
Configuration for a specific agent tool.
type BetaManagedAgentsAgentToolConfigParams = ...Configuration override for a specific tool within a toolset.
Configuration override for a specific tool within a toolset.
class BetaManagedAgentsAgentToolsetDefaultConfig: enabled; permission_policyResolved default configuration for agent tools.
Resolved default configuration for agent tools.
permission_policy: PermissionPolicyPermission policy for tool execution.
Permission policy for tool execution.
class BetaManagedAgentsAlwaysAllowPolicy: type: "always_allow"Tool calls are automatically approved without user confirmation.
Tool calls are automatically approved without user confirmation.
class BetaManagedAgentsAlwaysAskPolicy: type: "always_ask"Tool calls require user confirmation before execution.
Tool calls require user confirmation before execution.
class BetaManagedAgentsAutoPolicy: type: "auto"The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
class BetaManagedAgentsAgentToolsetDefaultConfigParams: enabled; permission_policyDefault configuration for all tools in a toolset.
Default configuration for all tools in a toolset.
class BetaManagedAgentsAgentToolset20260401: type: "agent_toolset_20260401"; configs; default_config
class BetaManagedAgentsAgentToolset20260401BashInput: command; restart; timeout_msInput payload for the bash tool of the
agent_toolset_20260401 toolset. All fields are optional;
a normal invocation supplies command, while restart=true
(with no command) reboots the runner-side bash session.
Input payload for the bash tool of the
agent_toolset_20260401 toolset. All fields are optional;
a normal invocation supplies command, while restart=true
(with no command) reboots the runner-side bash session.
Shell command to execute. Omit only when restart is true.
When true, restart the persistent bash session instead of
running a command. Subsequent calls without restart will
run against the fresh session.
timeout_ms: Optional[int]Per-call timeout in milliseconds. Defaults to the
runner-wide tool timeout when omitted or zero.
Per-call timeout in milliseconds. Defaults to the runner-wide tool timeout when omitted or zero.
class BetaManagedAgentsAgentToolset20260401EditInput: file_path; new_string; old_string; replace_allInput payload for the edit tool. Performs a string
replacement in the named file; by default old_string must
occur exactly once.
Input payload for the edit tool. Performs a string
replacement in the named file; by default old_string must
occur exactly once.
Path of the file to edit.
Replacement text.
Substring to find and replace.
When true, replace every occurrence of old_string
instead of requiring a unique match.
class BetaManagedAgentsAgentToolset20260401GlobInput: pattern; pathInput payload for the glob tool. Returns paths matching a
doublestar glob pattern, newest first.
Input payload for the glob tool. Returns paths matching a
doublestar glob pattern, newest first.
Doublestar glob pattern (e.g. **/*.go). Absolute patterns
are only permitted when the runner is configured to allow
them.
Optional directory root to search under. Defaults to the runner's working directory.
class BetaManagedAgentsAgentToolset20260401GrepInput: pattern; pathInput payload for the grep tool. Searches file contents for
a regular expression, returning matching lines.
Input payload for the grep tool. Searches file contents for
a regular expression, returning matching lines.
Regular expression to search for.
Optional directory root to search under. Defaults to the runner's working directory.
class BetaManagedAgentsAgentToolset20260401Params: type: "agent_toolset_20260401"; configs; default_configConfiguration for built-in agent tools. Use this to enable or disable groups of tools available to the agent.
Configuration for built-in agent tools. Use this to enable or disable groups of tools available to the agent.
class BetaManagedAgentsAgentToolset20260401ReadInput: file_path; view_rangeInput payload for the read tool. Reads file contents
relative to the runner's working directory (or absolute when
the runner permits).
Input payload for the read tool. Reads file contents
relative to the runner's working directory (or absolute when
the runner permits).
Path of the file to read.
view_range: Optional[List[int]]Optional [start_line, end_line] 1-indexed inclusive
range. When omitted the entire file is returned.
end_line of 0 or negative means "to end of file".
Optional [start_line, end_line] 1-indexed inclusive
range. When omitted the entire file is returned.
end_line of 0 or negative means "to end of file".
class BetaManagedAgentsAgentToolset20260401WriteInput: content; file_pathInput payload for the write tool. Writes (overwriting) the
entire file contents.
Input payload for the write tool. Writes (overwriting) the
entire file contents.
Full file contents to write.
Path of the file to write.
class BetaManagedAgentsAlwaysAllowPolicy: type: "always_allow"Tool calls are automatically approved without user confirmation.
Tool calls are automatically approved without user confirmation.
class BetaManagedAgentsAlwaysAskPolicy: type: "always_ask"Tool calls require user confirmation before execution.
Tool calls require user confirmation before execution.
class BetaManagedAgentsAnthropicSkill: type: "anthropic"; skill_id; versionA resolved Anthropic-managed skill.
A resolved Anthropic-managed skill.
class BetaManagedAgentsAnthropicSkillParams: type: "anthropic"; skill_id; versionAn Anthropic-managed skill.
An Anthropic-managed skill.
skill_id: strIdentifier of the Anthropic skill (e.g., "xlsx").
Identifier of the Anthropic skill (e.g., "xlsx").
version: Optional[str]Version to pin. Defaults to latest if omitted.
Version to pin. Defaults to latest if omitted.
class BetaManagedAgentsAutoPolicy: type: "auto"The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
class BetaManagedAgentsBashToolConfig: type: "bash"; enabled; name; permission_policyConfiguration for the bash tool.
Configuration for the bash tool.
class BetaManagedAgentsBashToolConfigParams: type; name; enabled; permission_policyConfiguration override for the bash tool.
Configuration override for the bash tool.
class BetaManagedAgentsCustomSkill: type: "custom"; skill_id; versionA resolved user-created custom skill.
A resolved user-created custom skill.
class BetaManagedAgentsCustomSkillParams: type: "custom"; skill_id; versionA user-created custom skill.
A user-created custom skill.
skill_id: strTagged ID of the custom skill (e.g., "skill_01XJ5...").
Tagged ID of the custom skill (e.g., "skill_01XJ5...").
version: Optional[str]Version to pin. Defaults to latest if omitted.
Version to pin. Defaults to latest if omitted.
class BetaManagedAgentsCustomTool: type: "custom"; description; input_schema; nameA custom tool as returned in API responses.
A custom tool as returned in API responses.
input_schema: BetaManagedAgentsCustomToolInputSchemaJSON Schema for custom tool input parameters.
JSON Schema for custom tool input parameters.
class BetaManagedAgentsCustomToolInputSchema: type: "object"; properties; requiredJSON Schema for custom tool input parameters.
JSON Schema for custom tool input parameters.
class BetaManagedAgentsCustomToolParams: type: "custom"; description; input_schema; nameA 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.
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.
description: strDescription of what the tool does, shown to the agent to help it decide when to use the tool.
Description of what the tool does, shown to the agent to help it decide when to use the tool.
input_schema: BetaManagedAgentsCustomToolInputSchemaJSON Schema defining the expected input parameters for the tool.
JSON Schema defining the expected input parameters for the tool.
name: strUnique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens.
Unique name for the tool. 1-128 characters; letters, digits, underscores, and hyphens.
class BetaManagedAgentsEditToolConfig: type: "edit"; enabled; name; permission_policyConfiguration for the edit tool.
Configuration for the edit tool.
class BetaManagedAgentsEditToolConfigParams: type; name; enabled; permission_policyConfiguration override for the edit tool.
Configuration override for the edit tool.
class BetaManagedAgentsEffortHigh: type: "high"High effort. Favors reasoning depth.
High effort. Favors reasoning depth.
class BetaManagedAgentsEffortLow: type: "low"Low effort. Favors latency over reasoning depth.
Low effort. Favors latency over reasoning depth.
class BetaManagedAgentsEffortMax: type: "max"Maximum effort. Favors reasoning depth over latency.
Maximum effort. Favors reasoning depth over latency.
class BetaManagedAgentsEffortMedium: type: "medium"Medium effort. Balances latency and reasoning depth.
Medium effort. Balances latency and reasoning depth.
class BetaManagedAgentsEffortXhigh: type: "xhigh"Extra-high effort. Not all models accept this level.
Extra-high effort. Not all models accept this level.
class BetaManagedAgentsGlobToolConfig: type: "glob"; enabled; name; permission_policyConfiguration for the glob tool.
Configuration for the glob tool.
class BetaManagedAgentsGlobToolConfigParams: type; name; enabled; permission_policyConfiguration override for the glob tool.
Configuration override for the glob tool.
class BetaManagedAgentsGrepToolConfig: type: "grep"; enabled; name; permission_policyConfiguration for the grep tool.
Configuration for the grep tool.
class BetaManagedAgentsGrepToolConfigParams: type; name; enabled; permission_policyConfiguration override for the grep tool.
Configuration override for the grep tool.
class BetaManagedAgentsMCPServerURLDefinition: type: "url"; name; urlURL-based MCP server connection as returned in API responses.
URL-based MCP server connection as returned in API responses.
class BetaManagedAgentsMCPToolConfig: enabled; name; permission_policyResolved configuration for a specific MCP tool.
Resolved configuration for a specific MCP tool.
class BetaManagedAgentsMCPToolConfigParams: name; enabled; permission_policyConfiguration override for a specific MCP tool.
Configuration override for a specific MCP tool.
class BetaManagedAgentsMCPToolset: type: "mcp_toolset"; configs; default_config; mcp_server_name
class BetaManagedAgentsMCPToolsetDefaultConfig: enabled; permission_policyResolved default configuration for all tools from an MCP server.
Resolved default configuration for all tools from an MCP server.
permission_policy: PermissionPolicyPermission policy for tool execution.
Permission policy for tool execution.
class BetaManagedAgentsAlwaysAllowPolicy: type: "always_allow"Tool calls are automatically approved without user confirmation.
Tool calls are automatically approved without user confirmation.
class BetaManagedAgentsAlwaysAskPolicy: type: "always_ask"Tool calls require user confirmation before execution.
Tool calls require user confirmation before execution.
class BetaManagedAgentsAutoPolicy: type: "auto"The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
The server decides each tool call individually: it judges, from the tool, its input, and the session content so far, whether the call is safe to execute or high-risk, and evaluates it to allow when judged safe and to deny when judged high-risk. A call the server cannot reach a judgement on evaluates to ask.
class BetaManagedAgentsMCPToolsetDefaultConfigParams: enabled; permission_policyDefault configuration for all tools from an MCP server.
Default configuration for all tools from an MCP server.
class BetaManagedAgentsMCPToolsetParams: type: "mcp_toolset"; mcp_server_name; configs; default_configConfiguration for tools from an MCP server defined in mcp_servers.
Configuration for tools from an MCP server defined in mcp_servers.
type BetaManagedAgentsModel = Union[Literal["claude-haiku-5-5", "claude-sonnet-5-5", "claude-opus-5-5", 14 more], str]The model that will power your agent.
The model that will power your agent.
See models for additional details and options.
class BetaManagedAgentsModelConfig: id; effort; inference_geo; speedModel identifier and configuration.
Model identifier and configuration.
class BetaManagedAgentsModelConfigParams: id; effort; inference_geo; speedAn object that defines additional configuration control over model use
An object that defines additional configuration control over model use
type BetaManagedAgentsMultiagentAdvisor = ...Whether the session's primary thread can consult an advisor model.
Whether the session's primary thread can consult an advisor model.
class BetaManagedAgentsMultiagentAdvisorEnabled: type: "enabled"; modelThe session's primary thread can consult model mid-turn.
The session's primary thread can consult model mid-turn.
The advisor model id.
class BetaManagedAgentsMultiagentAdvisorDisabled: type: "disabled"The agent has no advisor.
The agent has no advisor.
class BetaManagedAgentsMultiagentAdvisorDisabled: type: "disabled"The agent has no advisor.
The agent has no advisor.
class BetaManagedAgentsMultiagentAdvisorDisabledParams: type: "disabled"The agent has no advisor.
The agent has no advisor.
class BetaManagedAgentsMultiagentAdvisorEnabled: type: "enabled"; modelThe session's primary thread can consult model mid-turn.
The session's primary thread can consult model mid-turn.
The advisor model id.
class BetaManagedAgentsMultiagentAdvisorEnabledParams: type: "enabled"; modelThe session's primary thread can consult model mid-turn.
The session's primary thread can consult model mid-turn.
model: strA Claude model id. The model must be permitted as an advisor for this agent's model.
A Claude model id. The model must be permitted as an advisor for this agent's model.
type BetaManagedAgentsMultiagentAdvisorParams = ...Whether the session's primary thread can consult an advisor model.
Whether the session's primary thread can consult an advisor model.
class BetaManagedAgentsMultiagentAdvisorEnabledParams: type: "enabled"; modelThe session's primary thread can consult model mid-turn.
The session's primary thread can consult model mid-turn.
model: strA Claude model id. The model must be permitted as an advisor for this agent's model.
A Claude model id. The model must be permitted as an advisor for this agent's model.
class BetaManagedAgentsMultiagentAdvisorDisabledParams: type: "disabled"The agent has no advisor.
The agent has no advisor.
class BetaManagedAgentsMultiagentCoordinator: type: "coordinator"; agentsResolved coordinator topology with a concrete agent roster.
Resolved coordinator topology with a concrete agent roster.
class BetaManagedAgentsMultiagentCoordinatorParams: type: "coordinator"; agentsA coordinator topology: the session's primary thread orchestrates work by spawning session threads, each running an agent drawn from the agents roster.
A coordinator topology: the session's primary thread orchestrates work by spawning session threads, each running an agent drawn from the agents roster.
type BetaManagedAgentsMultiagentInlineAgents = ...Whether the agent can define inline agents. The agent defines an inline agent itself, in a workflow run's plan or when it spawns a session thread, and the inline agent is not saved.
Whether the agent can define inline agents. The agent defines an inline agent itself, in a workflow run's plan or when it spawns a session thread, and the inline agent is not saved.
class BetaManagedAgentsMultiagentInlineAgentsEnabled: type: "enabled"The agent can define inline agents.
The agent can define inline agents.
class BetaManagedAgentsMultiagentInlineAgentsDisabled: type: "disabled"The agent cannot define inline agents.
The agent cannot define inline agents.
class BetaManagedAgentsMultiagentInlineAgentsDisabled: type: "disabled"The agent cannot define inline agents.
The agent cannot define inline agents.
class BetaManagedAgentsMultiagentInlineAgentsDisabledParams: type: "disabled"The agent cannot define inline agents.
The agent cannot define inline agents.
class BetaManagedAgentsMultiagentInlineAgentsEnabled: type: "enabled"The agent can define inline agents.
The agent can define inline agents.
class BetaManagedAgentsMultiagentInlineAgentsEnabledParams: type: "enabled"The agent can define inline agents.
The agent can define inline agents.
type BetaManagedAgentsMultiagentInlineAgentsParams = ...Whether the agent can define inline agents. The agent defines an inline agent itself, in a workflow run's plan or when it spawns a session thread, and the inline agent is not saved.
Whether the agent can define inline agents. The agent defines an inline agent itself, in a workflow run's plan or when it spawns a session thread, and the inline agent is not saved.
class BetaManagedAgentsMultiagentInlineAgentsEnabledParams: type: "enabled"The agent can define inline agents.
The agent can define inline agents.
class BetaManagedAgentsMultiagentInlineAgentsDisabledParams: type: "disabled"The agent cannot define inline agents.
The agent cannot define inline agents.
type BetaManagedAgentsMultiagentPredefinedAgentParams = ...One agent in a predefined_agents list. It is an agent ID string, an agent reference with an optional version, or self for the agent that owns this configuration.
One agent in a predefined_agents list. It is an agent ID string, an agent reference with an optional version, or self for the agent that owns this configuration.
class BetaManagedAgentsAgentParams: type: "agent"; id; versionSpecification for an Agent. Provide a specific version or use the short-form agent="agent_id" for the most recent version
Specification for an Agent. Provide a specific version or use the short-form agent="agent_id" for the most recent version
id: strThe agent ID.
The agent ID.
version: Optional[int]The specific agent version to use. Omit to use the latest version. Must be at least 1 if specified.
The specific agent version to use. Omit to use the latest version. Must be at least 1 if specified.
class BetaManagedAgentsMultiagentSelfParams: type: "self"Sentinel roster entry meaning "the agent that owns this configuration". Resolved server-side to a concrete agent reference.
Sentinel roster entry meaning "the agent that owns this configuration". Resolved server-side to a concrete agent reference.
class BetaManagedAgentsMultiagentSelfParams: type: "self"Sentinel roster entry meaning "the agent that owns this configuration". Resolved server-side to a concrete agent reference.
Sentinel roster entry meaning "the agent that owns this configuration". Resolved server-side to a concrete agent reference.
type BetaManagedAgentsMultiagentSubagents = ...Whether the agent can spawn session threads.
Whether the agent can spawn session threads.
class BetaManagedAgentsMultiagentSubagentsEnabled: type: "enabled"; inline_agents; predefined_agentsThe agent can spawn session threads.
The agent can spawn session threads.
class BetaManagedAgentsMultiagentSubagentsDisabled: type: "disabled"The agent cannot spawn session threads.
The agent cannot spawn session threads.
class BetaManagedAgentsMultiagentSubagentsDisabled: type: "disabled"The agent cannot spawn session threads.
The agent cannot spawn session threads.
class BetaManagedAgentsMultiagentSubagentsDisabledParams: type: "disabled"The agent cannot spawn session threads.
The agent cannot spawn session threads.
class BetaManagedAgentsMultiagentSubagentsEnabled: type: "enabled"; inline_agents; predefined_agentsThe agent can spawn session threads.
The agent can spawn session threads.
class BetaManagedAgentsMultiagentSubagentsEnabledParams: type: "enabled"; inline_agents; predefined_agentsThe agent can spawn session threads. Each thread runs a predefined agent, which is a saved agent in predefined_agents, or an inline agent, which the agent defines when it spawns the thread and which is not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
The agent can spawn session threads. Each thread runs a predefined agent, which is a saved agent in predefined_agents, or an inline agent, which the agent defines when it spawns the thread and which is not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
type BetaManagedAgentsMultiagentSubagentsParams = ...Whether the agent can spawn session threads.
Whether the agent can spawn session threads.
class BetaManagedAgentsMultiagentSubagentsEnabledParams: type: "enabled"; inline_agents; predefined_agentsThe agent can spawn session threads. Each thread runs a predefined agent, which is a saved agent in predefined_agents, or an inline agent, which the agent defines when it spawns the thread and which is not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
The agent can spawn session threads. Each thread runs a predefined agent, which is a saved agent in predefined_agents, or an inline agent, which the agent defines when it spawns the thread and which is not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
class BetaManagedAgentsMultiagentSubagentsDisabledParams: type: "disabled"The agent cannot spawn session threads.
The agent cannot spawn session threads.
type BetaManagedAgentsMultiagentWorkflows = ...Whether the agent can start workflow runs.
Whether the agent can start workflow runs.
class BetaManagedAgentsMultiagentWorkflowsEnabled: type: "enabled"; inline_agents; predefined_agentsThe agent can start workflow runs.
The agent can start workflow runs.
class BetaManagedAgentsMultiagentWorkflowsDisabled: type: "disabled"The agent cannot start workflow runs.
The agent cannot start workflow runs.
class BetaManagedAgentsMultiagentWorkflowsDisabled: type: "disabled"The agent cannot start workflow runs.
The agent cannot start workflow runs.
class BetaManagedAgentsMultiagentWorkflowsDisabledParams: type: "disabled"The agent cannot start workflow runs.
The agent cannot start workflow runs.
class BetaManagedAgentsMultiagentWorkflowsEnabled: type: "enabled"; inline_agents; predefined_agentsThe agent can start workflow runs.
The agent can start workflow runs.
class BetaManagedAgentsMultiagentWorkflowsEnabledParams: type: "enabled"; inline_agents; predefined_agentsThe agent can start workflow runs. Each run follows a plan, a program that the agent writes. A plan can use predefined agents, which are the saved agents in predefined_agents, and inline agents, which it defines itself and which are not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
The agent can start workflow runs. Each run follows a plan, a program that the agent writes. A plan can use predefined agents, which are the saved agents in predefined_agents, and inline agents, which it defines itself and which are not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
type BetaManagedAgentsMultiagentWorkflowsParams = ...Whether the agent can start workflow runs.
Whether the agent can start workflow runs.
class BetaManagedAgentsMultiagentWorkflowsEnabledParams: type: "enabled"; inline_agents; predefined_agentsThe agent can start workflow runs. Each run follows a plan, a program that the agent writes. A plan can use predefined agents, which are the saved agents in predefined_agents, and inline agents, which it defines itself and which are not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
The agent can start workflow runs. Each run follows a plan, a program that the agent writes. A plan can use predefined agents, which are the saved agents in predefined_agents, and inline agents, which it defines itself and which are not saved. If inline_agents is disabled, predefined_agents must name at least one agent.
class BetaManagedAgentsMultiagentWorkflowsDisabledParams: type: "disabled"The agent cannot start workflow runs.
The agent cannot start workflow runs.
class BetaManagedAgentsMultiagent20261001: type: "multiagent_20261001"; advisor; subagents; workflowsResolved multiagent configuration with three members, each enabled or disabled on its own.
Resolved multiagent configuration with three members, each enabled or disabled on its own.
class BetaManagedAgentsMultiagent20261001Params: type: "multiagent_20261001"; advisor; subagents; workflowsMultiagent configuration with three members, each enabled or disabled on its own. On an update, if the agent's stored multiagent also has type multiagent_20261001, this configuration is merged into the stored one, level by level, instead of replacing it. A key that the update omits keeps its stored value. A key sent as null takes its default, on create as well, so "workflows": null enables workflows. An object sent with a type other than the stored one replaces the stored object, and the keys that it omits take their defaults. A predefined_agents list that is sent replaces the stored list. Every object that is sent needs its type, and an enabled advisor needs its model. Other validation applies to the merged result.
Multiagent configuration with three members, each enabled or disabled on its own. On an update, if the agent's stored multiagent also has type multiagent_20261001, this configuration is merged into the stored one, level by level, instead of replacing it. A key that the update omits keeps its stored value. A key sent as null takes its default, on create as well, so "workflows": null enables workflows. An object sent with a type other than the stored one replaces the stored object, and the keys that it omits take their defaults. A predefined_agents list that is sent replaces the stored list. Every object that is sent needs its type, and an enabled advisor needs its model. Other validation applies to the merged result.
class BetaManagedAgentsReadToolConfig: type: "read"; enabled; name; permission_policyConfiguration for the read tool.
Configuration for the read tool.
class BetaManagedAgentsReadToolConfigParams: type; name; enabled; permission_policyConfiguration override for the read tool.
Configuration override for the read tool.
class BetaManagedAgentsSessionThreadAgent: type: "agent"; id; description; ... # 7 moreResolved agent definition for a single session_thread. Snapshot of the agent at thread creation time. The multiagent roster is not repeated here; read it from Session.agent.
Resolved agent definition for a single session_thread. Snapshot of the agent at thread creation time. The multiagent roster is not repeated here; read it from Session.agent.
type BetaManagedAgentsSkillParams = ...Skill to load in the session container.
Skill to load in the session container.
class BetaManagedAgentsAnthropicSkillParams: type: "anthropic"; skill_id; versionAn Anthropic-managed skill.
An Anthropic-managed skill.
skill_id: strIdentifier of the Anthropic skill (e.g., "xlsx").
Identifier of the Anthropic skill (e.g., "xlsx").
version: Optional[str]Version to pin. Defaults to latest if omitted.
Version to pin. Defaults to latest if omitted.
class BetaManagedAgentsCustomSkillParams: type: "custom"; skill_id; versionA user-created custom skill.
A user-created custom skill.
skill_id: strTagged ID of the custom skill (e.g., "skill_01XJ5...").
Tagged ID of the custom skill (e.g., "skill_01XJ5...").
version: Optional[str]Version to pin. Defaults to latest if omitted.
Version to pin. Defaults to latest if omitted.
class BetaManagedAgentsURLMCPServerParams: type: "url"; name; urlURL-based MCP server connection.
URL-based MCP server connection.
name: strUnique name for this server, referenced by mcp_toolset configurations. 1-255 characters.
Unique name for this server, referenced by mcp_toolset configurations. 1-255 characters.
url: strEndpoint URL for the MCP server.
Endpoint URL for the MCP server.
class BetaManagedAgentsUserLocation: type: "approximate"; city; country; ... # 2 moreApproximate user location for search result localization.
Approximate user location for search result localization.
Location precision. Only "approximate" is supported.
city: Optional[str]City name.
City name.
Two-letter ISO 3166-1 country code, uppercase.
region: Optional[str]Region or state name.
Region or state name.
timezone: Optional[str]IANA timezone identifier, e.g. "America/Los_Angeles".
IANA timezone identifier, e.g. "America/Los_Angeles".
class BetaManagedAgentsWebFetchToolConfig: type: "web_fetch"; enabled; name; ... # 5 moreConfiguration for the web_fetch tool.
Configuration for the web_fetch tool.
class BetaManagedAgentsWebFetchToolConfigParams: type; name; allowed_domains; ... # 5 moreConfiguration override for the web_fetch tool.
Configuration override for the web_fetch tool.
class BetaManagedAgentsWebFetchURLSourceAll: type: "all"Every URL from this source may be fetched. This is the default.
Every URL from this source may be fetched. This is the default.
class BetaManagedAgentsWebFetchURLSourceExcept: type: "except"; toolsEvery tool's results contribute URLs that may be fetched, except the named tools' results.
Every tool's results contribute URLs that may be fetched, except the named tools' results.
tools: List[BetaManagedAgentsWebFetchURLSourceToolReference]The tools whose results do not contribute. Between 1 and 128 entries, each with a different name. An empty list is rejected; use "all" to leave out no tool's results.
The tools whose results do not contribute. Between 1 and 128 entries, each with a different name. An empty list is rejected; use "all" to leave out no tool's results.
Must be "tool_reference".
name: strName of the tool. Compared exactly, so upper and lower case letters are different.
Name of the tool. Compared exactly, so upper and lower case letters are different.
class BetaManagedAgentsWebFetchURLSourceNone: type: "none"This source contributes no URLs that may be fetched.
This source contributes no URLs that may be fetched.
class BetaManagedAgentsWebFetchURLSourceOnly: type: "only"; toolsOnly the named tools' results contribute URLs that may be fetched.
Only the named tools' results contribute URLs that may be fetched.
tools: List[BetaManagedAgentsWebFetchURLSourceToolReference]The tools whose results contribute. Between 1 and 128 entries, each with a different name. An empty list is rejected; use "none" to allow no tool's results.
The tools whose results contribute. Between 1 and 128 entries, each with a different name. An empty list is rejected; use "none" to allow no tool's results.
Must be "tool_reference".
name: strName of the tool. Compared exactly, so upper and lower case letters are different.
Name of the tool. Compared exactly, so upper and lower case letters are different.
type BetaManagedAgentsWebFetchURLSourceShorthand = Literal["all", "none"]String form of a url_sources value that has no field other than its type: "all" means {"type": "all"} and "none" means {"type": "none"}.
String form of a url_sources value that has no field other than its type: "all" means {"type": "all"} and "none" means {"type": "none"}.
type BetaManagedAgentsWebFetchURLSourceToolFilter = ...Which tools' results contribute URLs that may be fetched.
Which tools' results contribute URLs that may be fetched.
type BetaManagedAgentsWebFetchURLSourceToolFilterParams = Union[BetaManagedAgentsWebFetchURLSourceShorthand, BetaManagedAgentsWebFetchURLSourceToolFilter]Which tools' results contribute URLs that may be fetched. Accepts the string "all" or "none", or an object whose type is "all", "none", "only" or "except". Responses use the object form.
Which tools' results contribute URLs that may be fetched. Accepts the string "all" or "none", or an object whose type is "all", "none", "only" or "except". Responses use the object form.
Literal["all", "none"]
type BetaManagedAgentsWebFetchURLSourceToolFilter = ...Which tools' results contribute URLs that may be fetched.
Which tools' results contribute URLs that may be fetched.
class BetaManagedAgentsWebFetchURLSourceToolReference: type: "tool_reference"; nameNames one tool in an only or except list.
Names one tool in an only or except list.
Must be "tool_reference".
name: strName of the tool. Compared exactly, so upper and lower case letters are different.
Name of the tool. Compared exactly, so upper and lower case letters are different.
type BetaManagedAgentsWebFetchURLSourceUserInput = ...Whether URLs in the text of user messages may be fetched.
Whether URLs in the text of user messages may be fetched.
class BetaManagedAgentsWebFetchURLSourceAll: type: "all"Every URL from this source may be fetched. This is the default.
Every URL from this source may be fetched. This is the default.
class BetaManagedAgentsWebFetchURLSourceNone: type: "none"This source contributes no URLs that may be fetched.
This source contributes no URLs that may be fetched.
type BetaManagedAgentsWebFetchURLSourceUserInputParams = Union[BetaManagedAgentsWebFetchURLSourceShorthand, BetaManagedAgentsWebFetchURLSourceUserInput]Whether URLs in the text of user messages may be fetched. Accepts the string "all" or "none", or the object {"type": "all"} or {"type": "none"}. Responses use the object form.
Whether URLs in the text of user messages may be fetched. Accepts the string "all" or "none", or the object {"type": "all"} or {"type": "none"}. Responses use the object form.
Literal["all", "none"]
type BetaManagedAgentsWebFetchURLSourceUserInput = ...Whether URLs in the text of user messages may be fetched.
Whether URLs in the text of user messages may be fetched.
class BetaManagedAgentsWebFetchURLSourceAll: type: "all"Every URL from this source may be fetched. This is the default.
Every URL from this source may be fetched. This is the default.
class BetaManagedAgentsWebFetchURLSourceNone: type: "none"This source contributes no URLs that may be fetched.
This source contributes no URLs that may be fetched.
class BetaManagedAgentsWebFetchURLSources: client_tool_results; server_tool_results; user_inputWhich sources contribute URLs the web_fetch tool may fetch. A key that is null was not set and allows every URL from that source.
Which sources contribute URLs the web_fetch tool may fetch. A key that is null was not set and allows every URL from that source.
class BetaManagedAgentsWebFetchURLSourcesParams: client_tool_results; server_tool_results; user_inputWhich sources contribute URLs the web_fetch tool may fetch. When web_fetch is limited to URLs the conversation has already shown the model (in a user message, a custom tool's result, or an earlier web_search or web_fetch result), each key narrows one of those sources and defaults to "all". Setting all three keys to "none" is rejected.
Which sources contribute URLs the web_fetch tool may fetch. When web_fetch is limited to URLs the conversation has already shown the model (in a user message, a custom tool's result, or an earlier web_search or web_fetch result), each key narrows one of those sources and defaults to "all". Setting all three keys to "none" is rejected.
class BetaManagedAgentsWebSearchToolConfig: type: "web_search"; enabled; name; ... # 4 moreConfiguration for the web_search tool.
Configuration for the web_search tool.
class BetaManagedAgentsWebSearchToolConfigParams: type; name; allowed_domains; ... # 4 moreConfiguration override for the web_search tool.
Configuration override for the web_search tool.
class BetaManagedAgentsWriteToolConfig: type: "write"; enabled; name; permission_policyConfiguration for the write tool.
Configuration for the write tool.
class BetaManagedAgentsWriteToolConfigParams: type; name; enabled; permission_policyConfiguration override for the write tool.
Configuration override for the write tool.