iflytek/astron-agent · error · CustomException

KNOWLEDGE_PARAM_ERROR

KNOWLEDGE_PARAM_ERROR

Error message

docIds is empty

What it means

The knowledge-pro node requires a non-empty docIds list when repoType is CBG_RAG, because CBG RAG operates over an explicit set of documents. _check_cbg_rag_param raises a CustomException with KNOWLEDGE_PARAM_ERROR if docIds is an empty list for that repo type.

Solutions

1. Provide at least one document ID in the node's docIds field
2. If repoType shouldn't be CBG_RAG, switch repoType to a type that doesn't require docIds (e.g. a knowledge-base-scoped type)
3. Verify the upstream variable feeding docIds actually resolves to a non-empty list at runtime

Example fix

// before
{"repoType": "cbg_rag", "docIds": []}
// after
{"repoType": "cbg_rag", "docIds": ["doc-123", "doc-456"]}

if node_config["repoType"] == "cbg_rag" and not node_config.get("docIds"):
    raise ValueError("docIds is required when repoType is cbg_rag")

try:
    result = await node.async_execute(variable_pool, span)
except CustomException as e:
    if e.err_code == CodeEnum.KNOWLEDGE_PARAM_ERROR:
        # surface a user-facing config fix in the editor
        ...

When it happens

Trigger: Configuring a knowledge-pro node with repoType == 'cbg_rag' (RepoTypeEnum.CBG_RAG.value) while the docIds field is [] and calling execute.

Common situations: User picked the CBG RAG repository type in the workflow editor but never selected any documents; document IDs were bound to an upstream variable that resolved to empty; a template/JSON import omitted docIds.

Understand the failure class

Background: "must not be empty", "cannot be empty" — required-field validation errors across open-source libraries — this error's family across 41 libraries.

Related errors

• mcpServerId and mcpServerUrl cannot both be empty
• toolName cannot be empty
• WORKFLOW_VERSION_PUBLISH_FAILED
• 8008
• 8008

AI-assisted analysis of iflytek/astron-agent@5e758547a8 (2026-09-12). Data as JSON: /api/errors/606542bf3c331652. Report an issue: GitHub.

Appendix: source

Thrown at core/workflow/engine/nodes/knowledge_pro/knowledge_pro_node.py:95

    def run_f(self) -> WorkflowNodeExecutionStatus:
        """
        Get the failure execution status.

        :return: FAILED status for failed execution
        """
        return WorkflowNodeExecutionStatus.FAILED

    async def _check_cbg_rag_param(self) -> None:
        """
        Validate CBG RAG parameters.

        Ensures that docIds is not empty when using CBG_RAG repository type,
        as document IDs are required for CBG RAG operations.

        :raises CustomException: If docIds is empty for CBG_RAG repository type
        """
        if self.repoType == RepoTypeEnum.CBG_RAG.value and self.docIds == []:
            raise CustomException(
                err_code=CodeEnum.KNOWLEDGE_PARAM_ERROR, err_msg="docIds is empty"
            )

    async def execute(
        self, variable_pool: VariablePool, span: Span, **kwargs: Any
    ) -> NodeRunResult:
        """
        Execute the Knowledge Pro node operation.

        Performs RAG operations by querying knowledge repositories and generating
        responses using the configured LLM provider. Supports streaming responses
        and handles various error conditions.

        :param variable_pool: Pool of variables for the workflow execution
        :param span: Tracing span for observability
        :param kwargs: Additional keyword arguments including msg_or_end_node_deps
        :return: NodeRunResult containing execution status and outputs
        """

View on GitHub (pinned to 5e758547a8)