> ## Documentation Index > Fetch the complete documentation index at: https://docs.corti.ai/llms.txt > Use this file to discover all available pages before exploring further. # Update Agent by ID > This endpoint updates an existing agent. Only the fields provided in the request body will be updated; other fields will remain unchanged. ## OpenAPI ````yaml /agentic/auto-generated-openapi.yml patch /agents/{id} openapi: 3.1.1 info: title: Corti Agent API version: 0.1.0 description: > The Corti Agent API allows users to interact with the Corti Agent, enabling the creation and management of threads, sending messages, and receiving responses. servers: [] security: [] paths: /agents/{id}: patch: tags: - Agents summary: Update Agent by ID description: > This endpoint updates an existing agent. Only the fields provided in the request body will be updated; other fields will remain unchanged. operationId: agents_update parameters: - $ref: '#/components/parameters/Tenant-Name' - $ref: '#/components/parameters/AgentID' requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/AgentsUpdateAgent' responses: '200': description: Agent updated successfully. content: application/json: schema: $ref: '#/components/schemas/AgentsAgent' '400': description: > The request could not be processed, possibly due to missing required fields or invalid data. content: application/json: schema: $ref: '#/components/schemas/AgentsErrorResponse' '401': description: | Authentication is required to access this endpoint. content: application/json: schema: $ref: '#/components/schemas/AgentsErrorResponse' '404': description: Agent not found. content: application/json: schema: $ref: '#/components/schemas/AgentsErrorResponse' '422': description: | The request body failed validation. content: application/json: schema: $ref: '#/components/schemas/AgentsValidationErrorResponse' x-codeSamples: - lang: csharp label: C# .NET SDK source: > using Corti; var client = new CortiClient( "TENANT_NAME", CortiClientEnvironment.Eu, new CortiClientAuth.ClientCredentials("client_id", "client_secret") ); await client.Agents.UpdateAsync("12345678-90ab-cdef-gh12-34567890abc", new AgentsUpdateAgent()); - lang: javascript label: JavaScript SDK source: | import { CortiClient, CortiEnvironment } from "@corti/sdk"; const client = new CortiClient({ environment: CortiEnvironment.Eu, auth: { clientId: "YOUR_CLIENT_ID", clientSecret: "YOUR_CLIENT_SECRET" }, tenantName: "YOUR_TENANT_NAME" }); await client.agents.update("12345678-90ab-cdef-gh12-34567890abc"); components: parameters: Tenant-Name: name: Tenant-Name in: header description: >- Identifies a distinct entity within Corti's multi-tenant system. Ensures correct routing and authentication of the request. required: true example: base schema: type: string description: >- Identifies a distinct entity within Corti's multi-tenant system. Ensures correct routing and authentication of the request. example: base AgentID: name: id in: path required: true description: The identifier of the agent associated with the context. schema: type: string description: The identifier of the agent associated with the context. example: 12345678-90ab-cdef-gh12-34567890abc schemas: AgentsUpdateAgent: type: object properties: name: type: string description: The name of the agent. systemPrompt: type: string description: > The system prompt that defines the overall agents behavior and expectations. This field is optional as there is a default system orchestrator. description: type: string description: A brief description of the agent's capabilities. experts: type: array items: oneOf: - $ref: '#/components/schemas/AgentsCreateExpert' - $ref: '#/components/schemas/AgentsUpdateExpertReference' mcpServers: type: array description: > A list of MCP servers that the agent can call. If omitted, the agent can't call any MCP servers. items: $ref: '#/components/schemas/AgentsCreateMCPServer' AgentsAgent: type: object required: - id - name - description - systemPrompt properties: id: type: string description: The unique identifier of the agent. name: type: string description: The name of the agent. description: type: string description: A brief description of the agent's capabilities. systemPrompt: type: string description: >- The system prompt that defines the overall agents behavior and expectations. experts: type: array items: oneOf: - $ref: '#/components/schemas/AgentsExpert' - $ref: '#/components/schemas/AgentsExpertReference' mcpServers: type: array description: > A list of MCP servers that the agent can call. If omitted, the agent can't call any MCP servers. items: $ref: '#/components/schemas/AgentsMCPServer' AgentsErrorResponse: type: object required: - code - description properties: code: type: string description: | A machine-readable error code that identifies the type of error. description: type: string description: > A human-readable description of the error, providing more context about what went wrong. howToFix: type: string description: | A human-readable message describing how to fix the issue. details: type: object description: | An optional object containing additional details about the error. additionalProperties: true cause: type: object description: > An optional object containing the cause of the error, following the same structure as the parent error response. additionalProperties: true AgentsValidationErrorResponse: allOf: - $ref: '#/components/schemas/AgentsErrorResponse' - type: object properties: detail: type: array title: Detail items: $ref: '#/components/schemas/AgentsValidationError' AgentsCreateExpert: type: object required: - type - name - description properties: type: type: string enum: - new name: type: string description: The name of the expert. Must be unique. pattern: ^[a-zA-Z0-9_\.-]+$ description: type: string description: A brief description of the expert's capabilities. systemPrompt: type: string description: > Optional system prompt that defines the expert's behavior and expectations. mcpServers: type: array description: > A list of MCP servers that the expert can call. If omitted, the expert can't call any MCP Servers. items: $ref: '#/components/schemas/AgentsCreateMCPServer' AgentsUpdateExpertReference: description: >- An expert reference when updating an agent. The id is required to identify which expert to update, or reference. The expert must already exist. required: - id allOf: - $ref: '#/components/schemas/AgentsCreateExpertReference' AgentsCreateMCPServer: type: object required: - name - transportType - authorizationType - url properties: name: type: string description: Name of the MCP server. description: type: string description: A brief description of the MCP server's capabilities. transportType: type: string description: Type of transport used by the MCP server. enum: - stdio - streamable_http - sse authorizationType: type: string description: Type of authorization used by the MCP server. enum: - none - bearer - inherit - oauth2.0 authorizationScope: type: string description: OAuth2.0 authorization scope to request. url: type: string format: uri description: URL of the MCP server. redirectUrl: type: string format: uri description: Redirect URI for OAuth2.0 authorization. token: type: string description: Bearer token to be used in MCP client. AgentsExpert: type: object required: - id - type - name - description - systemPrompt properties: type: type: string enum: - expert id: type: string description: The unique identifier of the expert. name: type: string description: The name of the expert. Must be unique. pattern: ^[a-zA-Z0-9_\.-]+$ description: type: string description: A brief description of the expert's capabilities. systemPrompt: type: string description: > The system prompt that defines the expert's behavior and expectations. mcpServers: type: array description: > A list of MCP servers that the expert can call. If omitted, the expert can't call any MCP Servers. items: $ref: '#/components/schemas/AgentsMCPServer' AgentsExpertReference: type: object required: - type - id - name description: > A reference to an expert returned by the API. The expert's id and name are always provided. systemPrompt is included only when a registry expert was created with a custom system prompt. properties: type: type: string enum: - reference id: type: string description: The unique identifier of the expert. name: type: string description: The name of the expert. pattern: ^[a-zA-Z0-9_\.-]+$ systemPrompt: type: string description: > Optional. Additional system instructions appended to the default system prompt when creating an expert from the registry, extending the expert's behavior. resolvedConfig: type: object description: > The fully resolved configuration for this expert: schema defaults from the registry expert deep-merged with any `config` override supplied at creation. Present only when the source registry expert defined a `configSchema`. additionalProperties: true AgentsMCPServer: type: object required: - id - name - transportType - authorizationType - url properties: id: type: string description: Unique identifier for the MCP server. name: type: string description: Name of the MCP server. transportType: type: string description: Type of transport used by the MCP server. enum: - stdio - streamable_http - sse authorizationType: type: string description: Type of authorization used by the MCP server. enum: - none - bearer - inherit - oauth2.0 authorizationScope: type: string description: OAuth2.0 authorization scope to request. url: type: string format: uri description: URL of the MCP server. redirectUrl: type: - string - 'null' format: uri description: Redirect URI for OAuth2.0 authorization. AgentsValidationError: properties: msg: type: string title: Message type: type: string title: Error Type reason: type: string title: Reason howToFix: type: string title: How to Fix errors: type: array items: type: object properties: location: type: string title: Location reason: type: string title: Reason title: Errors type: object required: - msg - type - reason - howToFix title: AgentsValidationError AgentsCreateExpertReference: type: object required: - type description: > A reference to a registry expert when creating an agent, either id or name must be provided. If both are passed, the id will be used. properties: type: type: string enum: - reference id: type: string description: The unique identifier of the expert. name: type: string description: The name of the expert. pattern: ^[a-zA-Z0-9_\.-]+$ systemPrompt: type: string description: > Optional. Additional system instructions appended to the default system prompt when creating an expert from the registry, extending the expert's behavior. config: type: object description: > Optional configuration override for the registry expert. Values provided here are deep-merged with the schema defaults declared on the registry expert and validated against its `configSchema`. Ignored when the registry expert has no schema. additionalProperties: true ````