> ## Documentation Index > Fetch the complete documentation index at: https://docs.streamkap.com/llms.txt > Use this file to discover all available pages before exploring further. # Test Run Agent > Run an agent against a single sample record without deploying to Flink. Synchronous, hard-capped at 15s wall-clock. Calls the LLM provider directly + dispatches MCP and HTTP tools live; transform tools and memory are intentionally not exercised (the Flink runtime owns them). Per-tenant rate limit: shares the validate-llm bucket (30 req/min/tenant) -> 429. Provider failures (timeouts, schema errors, unsupported provider) land as a structured ``error`` object on the envelope (HTTP still 200) so the FE can render the partial timeline alongside the error context. Only the input-token cap raises (422) - at that point we can't even start the run. ## OpenAPI ````yaml /openapi/openapi.json post /agents/test-run openapi: 3.1.0 info: title: Streamkap REST API description: >- The Streamkap REST API allows you to programmatically manage your CDC pipelines, sources, destinations, transforms, and more. Authenticate using a bearer token obtained from the [Access Token](/api-reference/auth/access-token) endpoint with your API client credentials. contact: name: Streamkap Support url: https://streamkap.com/ email: support@streamkap.com license: name: Proprietary version: 2.0.0 servers: - url: https://api.streamkap.com description: Production security: [] tags: - name: Agents description: Create, deploy, and manage Flink-based AI agents. - name: Agents Observability description: >- Read-only views over the MCP audit log: agent rollup, sessions, spans, histograms, external-agent verify. - name: Alerts description: Manage alert subscribers, preferences, and notification credentials. - name: Authentication description: Obtain and refresh access tokens, manage client credentials and roles. - name: Billing description: Retrieve usage metrics, summaries, and export billing data. - name: Consumer Groups description: List, inspect, and reset Kafka consumer group offsets. - name: Dashboard description: Retrieve organisation-level statistics and overview data. - name: Destinations description: Create, configure, and manage data destinations and their lifecycle. - name: Kafka Access description: Manage Kafka users and their access permissions. - name: Logs description: Query and summarise system logs. - name: Pipelines description: Create, configure, and manage CDC pipelines and their lifecycle. - name: Project Keys description: Manage project keys that bundle API credentials and optional Kafka access. - name: Schema Registry description: Browse schema subjects, versions, and retrieve schema definitions. - name: Services description: View and switch service metadata. - name: Sources description: >- Create, configure, and manage data sources, snapshots, and their lifecycle. - name: Tags description: Create, update, and manage resource tags for organisation. - name: Topics description: Browse topic details, statistics, configurations, metrics, and messages. - name: Transforms description: >- Create, deploy, and manage data transforms, unit tests, and implementation details. paths: /agents/test-run: post: tags: - Agents summary: Test Run Agent description: |- Run an agent against a single sample record without deploying to Flink. Synchronous, hard-capped at 15s wall-clock. Calls the LLM provider directly + dispatches MCP and HTTP tools live; transform tools and memory are intentionally not exercised (the Flink runtime owns them). Per-tenant rate limit: shares the validate-llm bucket (30 req/min/tenant) -> 429. Provider failures (timeouts, schema errors, unsupported provider) land as a structured ``error`` object on the envelope (HTTP still 200) so the FE can render the partial timeline alongside the error context. Only the input-token cap raises (422) - at that point we can't even start the run. operationId: testRunAgentOnSampleRecord requestBody: content: application/json: schema: $ref: '#/components/schemas/TestRunRequest' required: true responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/TestRunResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - FronteggHTTPAuthentication: [] components: schemas: TestRunRequest: properties: config: $ref: '#/components/schemas/CreateAgentConfigRequest' description: Full agent config under test sampleRecord: additionalProperties: true type: object title: Samplerecord description: One Kafka-shaped record to feed the agent agentId: anyOf: - type: string maxLength: 64 - type: 'null' title: Agentid description: >- Resolves an empty / masked ``apiKey`` against the stored agent's encrypted blob. Post-cutover the wizard sends ``apiKey: ''`` for every existing-agent test-run; this lets the BE swap in the stored key for legacy inline-key agents that were saved before the Connection cutover (no ``llmConnectionId`` to resolve via). savedConnectionId: anyOf: - type: string maxLength: 64 - type: 'null' title: Savedconnectionid description: Mirror of the ValidateLlmRequest.savedConnectionId path. skipTools: type: boolean title: Skiptools description: >- When true, every MCP / HTTP tool dispatch returns a canned {skipped: true} payload to the LLM. Use to iterate on prompt shape without burning real tool calls. default: false additionalProperties: false type: object required: - config - sampleRecord title: TestRunRequest description: |- Request body for ``POST /agents/test-run``. Carries a full agent config + a sample record. The BE runs the record through the agent (LLM + tools) without touching Flink and returns the timeline so the FE can show what happened. TestRunResponse: properties: steps: items: $ref: '#/components/schemas/TestRunStep' type: array title: Steps finalOutput: title: Finaloutput stoppedReason: type: string enum: - completed - max_iterations - error - timeout title: Stoppedreason iterationsUsed: type: integer title: Iterationsused tokensIn: type: integer title: Tokensin tokensOut: type: integer title: Tokensout durationMs: type: integer title: Durationms approximations: items: type: string type: array title: Approximations description: >- Free-form list of caveats, e.g. 'memory_excluded', 'transforms_skipped'. suggestions: items: $ref: '#/components/schemas/TestRunSuggestion' type: array title: Suggestions description: >- Deterministic hints (max 4) computed from the agent config + run result. Empty when the run produced a useful structured output. Each suggestion has a stable ``kind`` the FE can map to icon / copy. warnings: items: $ref: '#/components/schemas/TestRunWarning' type: array title: Warnings description: >- Static config-vs-record checks (e.g. memory keyField missing on the sample) that the Flink runtime would silently swallow at deploy. Empty when nothing is amiss. outputSchemaCheck: anyOf: - $ref: '#/components/schemas/TestRunSchemaCheck' - type: 'null' description: >- Field-level diff between the declared ``output.schema`` and the LLM's ``finalOutput``. Null when no schema was declared OR the run errored / produced a non-structured (text envelope) result. ``ok=true`` means every declared field is present with a compatible type. systemPrompt: anyOf: - type: string - type: 'null' title: Systemprompt description: >- Composed system prompt the LLM actually received - base template + auto-injected schema fields + tool descriptions + customInstructions. Identical to what Flink sends at deploy time (built by the same ``build_system_prompt``). Null when the run errored before reaching the LLM (e.g. ``llm_resolution_error``). userMessage: anyOf: - type: string - type: 'null' title: Usermessage description: >- First user message dispatched to the LLM, after the ``prompts.user`` template wraps the (input-fields-filtered) sample record. Lets the test pane show the exact pair of strings the LLM saw on iteration 1. Null when the run errored before message construction. error: anyOf: - $ref: '#/components/schemas/TestRunError' - type: 'null' additionalProperties: false type: object required: - steps - stoppedReason - iterationsUsed - tokensIn - tokensOut - durationMs title: TestRunResponse description: Response body for ``POST /agents/test-run``. HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError CreateAgentConfigRequest: properties: name: type: string maxLength: 200 minLength: 1 title: Name description: Agent display name description: anyOf: - type: string maxLength: 500 - type: 'null' title: Description description: Human-readable context for the agent enabled: type: boolean title: Enabled description: >- Kill switch. When false, the Flink runtime accepts the deploy but stops emitting output — used by support to disable a runaway tenant agent without a full cancel/redeploy. default: true agentType: $ref: '#/components/schemas/AgentTypeEnum' description: >- Agent type — derived from the tool list post-validation; client-supplied value is overwritten default: workflow input: $ref: '#/components/schemas/AgentInputConfig' output: $ref: '#/components/schemas/AgentOutputConfig' llm: $ref: '#/components/schemas/AgentLlmConfig' prompts: $ref: '#/components/schemas/AgentPromptConfig' mcpServer: anyOf: - $ref: '#/components/schemas/AgentMcpServerConfig' - type: 'null' description: Shared MCP server config for all MCP tools tools: items: $ref: '#/components/schemas/AgentToolConfig' type: array title: Tools description: Agent tools (MCP, HTTP, transform) memory: anyOf: - $ref: '#/components/schemas/AgentMemoryConfig' - type: 'null' description: Memory config (short-term + long-term) knowledgeBases: items: $ref: '#/components/schemas/AgentKnowledgeBaseRef' type: array maxItems: 10 title: Knowledgebases description: Knowledge bases this agent can query at runtime for RAG processing: $ref: '#/components/schemas/AgentProcessingConfig' additionalProperties: false type: object required: - name - input - output - llm - prompts title: CreateAgentConfigRequest description: Request body for POST /agents/config - create a config-based agent. TestRunStep: properties: kind: type: string enum: - llm - tool title: Kind model: anyOf: - type: string - type: 'null' title: Model tokensIn: anyOf: - type: integer - type: 'null' title: Tokensin tokensOut: anyOf: - type: integer - type: 'null' title: Tokensout stopReason: anyOf: - type: string - type: 'null' title: Stopreason text: anyOf: - type: string - type: 'null' title: Text toolCalls: anyOf: - items: additionalProperties: true type: object type: array - type: 'null' title: Toolcalls toolName: anyOf: - type: string - type: 'null' title: Toolname toolType: anyOf: - type: string - type: 'null' title: Tooltype request: anyOf: - additionalProperties: true type: object - type: 'null' title: Request result: anyOf: - additionalProperties: true type: object - type: 'null' title: Result skipped: anyOf: - type: boolean - type: 'null' title: Skipped skippedReason: anyOf: - type: string - type: 'null' title: Skippedreason durationMs: anyOf: - type: integer - type: 'null' title: Durationms error: anyOf: - type: string - type: 'null' title: Error additionalProperties: false type: object required: - kind title: TestRunStep description: One row in the test-run timeline. Either an LLM call or a tool dispatch. TestRunSuggestion: properties: kind: type: string enum: - missing_output_schema - tools_not_invoked - near_empty_response - skip_instruction_may_apply - max_tokens_truncated title: Kind message: type: string title: Message additionalProperties: false type: object required: - kind - message title: TestRunSuggestion description: |- A deterministic, config-derived hint for improving a test-run result. Surfaced when a run completes successfully but the output is empty, near-empty, or otherwise unhelpful AND the cause is reliably identifiable from the agent config alone (no LLM-grade judgment). The FE renders these as a "Suggestions" panel under the result. ``kind`` is a stable enum the FE can map to icon / link copy. The ``message`` is a default rendering for clients that don't know the kind. Adding new kinds is non-breaking - the FE falls back to the message string for any unknown ``kind``. TestRunWarning: properties: kind: type: string enum: - memory_key_field_missing - input_fields_not_in_record - max_tokens_clamped title: Kind message: type: string title: Message additionalProperties: false type: object required: - kind - message title: TestRunWarning description: |- Static, pre-flight warning surfaced before / alongside the run. Distinct from ``error`` (which represents a run failure) and from ``suggestions`` (post-hoc empty-output hints). Warnings are deterministic config-vs-record checks (e.g. memory keyField missing) that the Flink runtime would silently swallow at deploy time. TestRunSchemaCheck: properties: ok: type: boolean title: Ok missing: items: type: string type: array title: Missing description: Fields declared in schema but absent from finalOutput extra: items: type: string type: array title: Extra description: Fields present in finalOutput but not declared typeMismatches: items: $ref: '#/components/schemas/TestRunSchemaTypeMismatch' type: array title: Typemismatches description: Fields whose value types diverge from the declared schema additionalProperties: false type: object required: - ok title: TestRunSchemaCheck description: >- Diff between the declared ``output.schema`` and the LLM's ``finalOutput``. Computed only when (a) the agent declared a schema and (b) the run produced a structured (non-text-envelope) final output. The Flink runtime DLQs records that fail this check at deploy time; surfacing the same comparison in test mode closes the audit's "silent output-schema bypass" gap. TestRunError: properties: kind: type: string title: Kind description: >- One of: llm_resolution_error, llm_provider_error, timeout, unsupported_provider, http_connection_resolution_error, … message: type: string title: Message additionalProperties: false type: object required: - kind - message title: TestRunError description: Structured error envelope on a failed test-run. ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError AgentTypeEnum: type: string enum: - workflow - react title: AgentTypeEnum AgentInputConfig: properties: topicPattern: type: string maxLength: 1024 minLength: 1 title: Topicpattern description: Regex pattern for input topics, e.g. '^(orders)$' inputSerialization: $ref: '#/components/schemas/AgentSerializationEnum' description: Input deserialization format default: JsonConfluent fields: anyOf: - items: type: string type: array - type: 'null' title: Fields description: Column-level filter - only these fields sent to LLM createTableSQL: anyOf: - type: string maxLength: 4000 - type: 'null' title: Createtablesql description: Flink SQL CREATE TABLE (required if filterSQL is set) filterSQL: anyOf: - type: string maxLength: 4000 - type: 'null' title: Filtersql description: WHERE clause predicate for row-level filtering additionalProperties: false type: object required: - topicPattern title: AgentInputConfig description: Input configuration for the agent. AgentOutputConfig: properties: topic: type: string maxLength: 249 minLength: 1 title: Topic description: Fixed output topic name deadLetterTopic: anyOf: - type: string maxLength: 249 - type: 'null' title: Deadlettertopic description: DLQ topic name outputSerialization: $ref: '#/components/schemas/AgentSerializationEnum' description: Output serialization format default: JsonConfluent schema: additionalProperties: $ref: '#/components/schemas/AgentOutputSchemaTypeEnum' type: object title: Schema description: 'Output schema: {field_name: TYPE_STRING}' additionalProperties: false type: object required: - topic title: AgentOutputConfig description: Output configuration for the agent. AgentLlmConfig: properties: provider: $ref: '#/components/schemas/AgentLlmProviderEnum' description: LLM provider model: type: string minLength: 1 title: Model description: Model name, e.g. claude-sonnet-4-20250514 apiKey: type: string maxLength: 512 title: Apikey description: >- API key - use '${SECRET:ENV_VAR_NAME}' for env var resolution. Empty when llmConnectionId is set. default: '' llmConnectionId: anyOf: - type: string maxLength: 100 - type: 'null' title: Llmconnectionid description: >- Tenant-scoped reference to ``agent_connections.llmConnections[].id``. When set with an empty ``apiKey``, the BE resolves the real key server-side at deploy and at test-run. baseUrl: anyOf: - type: string - type: 'null' title: Baseurl description: Base URL for OpenAI-compatible proxies temperature: anyOf: - type: number maximum: 2 minimum: 0 - type: 'null' title: Temperature description: Sampling temperature (null = matrix-skipped) maxTokens: anyOf: - type: integer maximum: 100000 minimum: 1 - type: 'null' title: Maxtokens description: Max output tokens (null = matrix-skipped) timeout: anyOf: - type: integer maximum: 600 minimum: 1 - type: 'null' title: Timeout description: Request timeout in seconds (null = matrix-skipped) reasoningEffort: anyOf: - type: string enum: - none - minimal - low - medium - high - xhigh - type: 'null' title: Reasoningeffort description: >- OpenAI / GPT-5 reasoning effort. The accepted set is per-model: the matrix in app/utils/llm_capabilities.py rejects values not in caps.reasoning_effort_values for the chosen provider+model. thinkingBudgetTokens: anyOf: - type: integer maximum: 64000 minimum: 0 - type: 'null' title: Thinkingbudgettokens description: >- Anthropic Claude 4 extended-thinking budget (tokens). Non-null enables thinking. ollamaThink: anyOf: - type: boolean - type: 'null' title: Ollamathink description: Ollama 'think' toggle for reasoning models maxRetries: anyOf: - type: integer maximum: 20 minimum: 0 - type: 'null' title: Maxretries description: >- Universal — max retry attempts on LLM call. Accepted by every provider. region: anyOf: - type: string maxLength: 64 - type: 'null' title: Region description: >- Bedrock: AWS region (e.g. us-east-1). Ignored for non-bedrock providers. strict: anyOf: - type: boolean - type: 'null' title: Strict description: 'openai-responses: enable JSON-schema strict mode.' store: anyOf: - type: boolean - type: 'null' title: Store description: 'openai-responses: server-side response storage flag.' instructions: anyOf: - type: string maxLength: 8192 - type: 'null' title: Instructions description: >- openai-responses: system-level instructions passed as a top-level Responses param. additionalKwargs: anyOf: - additionalProperties: true type: object - type: 'null' title: Additionalkwargs description: >- openai-responses / azure-openai: free-form extra request params forwarded verbatim by the runtime. apiVersion: anyOf: - type: string maxLength: 32 - type: 'null' title: Apiversion description: >- azure-openai: Azure OpenAI API version (e.g. '2024-02-01'). Required for azure-openai. azureEndpoint: anyOf: - type: string maxLength: 512 - type: 'null' title: Azureendpoint description: >- azure-openai: Azure resource endpoint (e.g. https://.openai.azure.com). Required. azureUrlPathMode: anyOf: - type: string enum: - AUTO - LEGACY - UNIFIED - type: 'null' title: Azureurlpathmode description: >- azure-openai: URL path resolution — AUTO / LEGACY / UNIFIED. Optional. additionalProperties: false type: object required: - provider - model title: AgentLlmConfig description: |- Inline LLM configuration embedded on the agent. Carries provider, credentials, model, and tuning controls as a single blob. The FE prefills the form from a saved :class:`AgentLlmConnection` (which now carries default model + tuning + reasoning controls) and the user can override per-agent before save; the resulting blob is stored verbatim and shipped to the Flink runtime at deploy. Two credential shapes (XOR, mirrors :class:`ValidateLlmRequest`): 1. **Inline** - ``apiKey`` is filled directly. The BE KMS-encrypts it at save and decrypts it at deploy / test-run. 2. **Linked** - ``llmConnectionId`` references a row in ``agent_connections.llmConnections`` and ``apiKey`` is left empty. The BE resolves the real key server-side at deploy and at test-run via :func:`agents_service.resolve_saved_llm_credentials`, so the browser never needs to hold (or re-paste) the stored plaintext when editing an existing agent. ``provider == ollama`` accepts an empty key without a link (the runtime is local and unauthenticated). AgentPromptConfig: properties: system: type: string maxLength: 8000 title: System description: Assembled system prompt (auto-generated by backend) default: '' customInstructions: type: string maxLength: 8000 title: Custominstructions description: User-provided instructions appended to the auto-generated prompt default: '' user: type: string maxLength: 2000 title: User description: >- User prompt template - {input_json} is replaced with the input record default: {input_json} additionalProperties: false type: object title: AgentPromptConfig description: >- Prompt configuration for the agent. The final system prompt is auto-assembled by ``build_system_prompt()`` from the agent type template + output schema + tool descriptions + custom_instructions. The ``system`` field stores the assembled result (set by the backend, not the user). AgentMcpServerConfig: properties: projectKeyId: anyOf: - type: string maxLength: 128 - type: 'null' title: Projectkeyid description: >- ID of an agentic-enabled Project Key. When set, deploy-time resolves the Streamkap MCP serverUrl + auth header from the PK row. Mutually exclusive with serverUrl - one or the other, never both. serverUrl: anyOf: - type: string maxLength: 2048 minLength: 1 - type: 'null' title: Serverurl description: >- External MCP server URL. Required when projectKeyId is unset. Server-stamped from the PK row when projectKeyId is set. headers: additionalProperties: type: string type: object maxProperties: 32 title: Headers description: Auth headers for MCP server additionalProperties: false type: object title: AgentMcpServerConfig description: >- Top-level MCP server config shared across all MCP tools. Two shapes (hard cutover): - **Streamkap MCP** (internal toolbelt): ``projectKeyId`` set, ``serverUrl`` and ``headers`` resolved server-side at deploy time from the agentic-enabled Project Key. The legacy "saved Streamkap MCP card" path (match-by-serverUrl against ``agent_connections.mcpConnections[source==streamkap]``) is gone. - **External MCP**: ``serverUrl`` + ``headers`` set inline; ``projectKeyId`` omitted. Headers are still composed at deploy time from the matching saved external connection (bearer / custom-header). AgentToolConfig: properties: name: type: string maxLength: 100 minLength: 1 title: Name description: Tool name (must be unique) type: $ref: '#/components/schemas/AgentToolTypeEnum' description: 'Tool type: http, transform, or mcp' description: type: string maxLength: 2000 title: Description description: Tool description for the LLM default: '' config: additionalProperties: true type: object title: Config description: Type-specific configuration parameters: items: $ref: '#/components/schemas/AgentToolParameterConfig' type: array title: Parameters description: Tool parameters the LLM can provide additionalProperties: false type: object required: - name - type title: AgentToolConfig description: |- A single tool definition in the agent config. ``name`` and ``description`` are concatenated into the assembled system prompt by ``build_system_prompt()`` and shipped to Flink as part of the ``--agentConfig`` CLI argument. Length caps here keep the worst-case serialized agentConfig within the Flink CLI ~64KB single-arg budget even with many tools configured. AgentMemoryConfig: properties: keyField: anyOf: - type: string - type: 'null' title: Keyfield description: >- Field to partition memory by (e.g. 'customer_id'). Default: topic key shortTerm: $ref: '#/components/schemas/AgentShortTermMemoryConfig' description: Short-term conversation memory longTerm: $ref: '#/components/schemas/AgentLongTermMemoryConfig' description: Long-term vector store memory additionalProperties: false type: object title: AgentMemoryConfig description: Memory configuration for the agent. AgentKnowledgeBaseRef: properties: id: type: string maxLength: 64 minLength: 1 title: Id description: Knowledge base entity ID name: type: string maxLength: 200 title: Name description: Display name (denormalized for UI) default: '' additionalProperties: false type: object required: - id title: AgentKnowledgeBaseRef description: >- Reference to a Knowledge Base that an agent can query at runtime for RAG. Stored on the agent config. At deploy time, the BE resolves each ref to the KB's vector store + embedding credentials and injects them into the Flink agentConfig JSON. AgentProcessingConfig: properties: parallelism: type: integer maximum: 128 minimum: 1 title: Parallelism description: Parallelism default: 1 checkpointIntervalMin: type: integer maximum: 60 minimum: 1 title: Checkpointintervalmin description: Checkpoint interval in minutes default: 5 maxIterations: type: integer maximum: 25 minimum: 1 title: Maxiterations description: Max tool-call iterations per record (react type only) default: 10 maxTokensPerHour: type: integer minimum: 0 title: Maxtokensperhour description: >- Per-TaskManager rolling-hour token budget. 0 = unlimited. When exceeded, records pass through with a `_budget_exceeded: true` marker. default: 0 additionalProperties: false type: object title: AgentProcessingConfig description: Processing configuration for the agent. TestRunSchemaTypeMismatch: properties: field: type: string title: Field expected: type: string title: Expected description: Declared schema type, e.g. STRING / DOUBLE actual: type: string title: Actual description: Inferred type of the value the LLM returned additionalProperties: false type: object required: - field - expected - actual title: TestRunSchemaTypeMismatch description: >- One field-level mismatch between declared output schema and the LLM's response. AgentSerializationEnum: type: string enum: - JsonConfluent - AvroConfluent title: AgentSerializationEnum AgentOutputSchemaTypeEnum: type: string enum: - STRING - DOUBLE - INT32 - INT64 - BOOLEAN title: AgentOutputSchemaTypeEnum AgentLlmProviderEnum: type: string enum: - anthropic - openai - openai-responses - ollama - azure - azure-openai - bedrock - qwen - openai-compatible title: AgentLlmProviderEnum description: >- Unified LLM provider enum. A single ``AgentLlmConnection`` row carries one provider and a set of capabilities (chat / embedding). ``PROVIDER_CAPABILITIES`` below pins which capabilities each provider can serve — picked by the FE Connections drawer and re-validated server-side on every write. AgentToolTypeEnum: type: string enum: - http - transform - mcp title: AgentToolTypeEnum AgentToolParameterConfig: properties: name: type: string maxLength: 100 minLength: 1 title: Name description: Parameter name type: type: string maxLength: 50 title: Type description: Parameter type (string, number, boolean) default: string description: type: string maxLength: 2000 title: Description description: Parameter description for the LLM default: '' additionalProperties: false type: object required: - name title: AgentToolParameterConfig description: A single parameter for a tool. AgentShortTermMemoryConfig: properties: enabled: type: boolean title: Enabled description: Enable short-term memory default: false ttlMs: type: integer minimum: 0 title: Ttlms description: 'Time-to-live in milliseconds (default: 1 hour)' default: 3600000 maxEntries: type: integer maximum: 100 minimum: 1 title: Maxentries description: Max conversation entries per key default: 20 additionalProperties: false type: object title: AgentShortTermMemoryConfig description: Short-term (conversation) memory config. AgentLongTermMemoryConfig: properties: enabled: type: boolean title: Enabled description: Enable long-term memory default: false vectorStoreConnectionId: anyOf: - type: string maxLength: 64 - type: 'null' title: Vectorstoreconnectionid description: >- Reference to a saved AgentVectorStoreConnection. Resolved at deploy time. namespace: anyOf: - type: string maxLength: 200 - type: 'null' title: Namespace description: >- Per-agent namespace override (falls back to connection's defaultNamespace) destinationId: anyOf: - type: string maxLength: 64 - type: 'null' title: Destinationid description: >- (Legacy) ObjectId of a Pinecone destination connector. Prefer vectorStoreConnectionId. additionalProperties: false type: object title: AgentLongTermMemoryConfig description: >- Long-term (vector store) memory config. References a saved ``AgentVectorStoreConnection`` by id. At deploy time the resolver reads the connection's ``apiKey`` + ``endpoint`` and composes the inline ``vectorStore`` block the Java runtime expects. Legacy agents may still carry ``destinationId`` (a Pinecone destination connector from the old flow). The deploy resolver handles both: if ``vectorStoreConnectionId`` is set it wins; otherwise it falls back to the legacy ``destinationId`` path. New agents created through the FE always use ``vectorStoreConnectionId``. securitySchemes: FronteggHTTPAuthentication: type: http scheme: bearer ````