ReportAgentExecutionOutputs

Reports outputs for an agent execution.

This method allows agents to report key-value outputs during execution. Similar to task execution outputs, but with typed values for structured data. Outputs are merged with any existing outputs. The API automatically infers the type (int, float, bool, or string) from the value.

Use this method to:

• Record execution outcomes and metrics
• Track test results, coverage, or success indicators
• Store any execution-related data as key-value pairs

Type inference rules:

• "true" or "false" → bool
• Valid integer (e.g., "42", "-10") → int
• Valid float (e.g., "85.5", "3.14") → float
• Everything else → string

Validation limits (matching task execution outputs):

• Keys: 1-128 characters
• String values: max 4096 characters

Examples

• Report execution outputs (types will be inferred from string_value):

  agentExecutionId: "6fa1a3c7-fbb7-49d1-ba56-1890dc7c4c35"
  outputs:
    tests_passed:
      stringValue: "42"        # inferred as int
    coverage:
      stringValue: "85.5"      # inferred as float
    success:
      stringValue: "true"      # inferred as bool
    message:
      stringValue: "All tests passed"  # stored as string