Count of generated elements by category for this phase.
Provides the number of key artifacts generated during the phase execution. Each phase produces different types of elements:
Keys represent element type names (e.g., "actors", "operations") and values indicate the count of each element type generated, offering insights into the scope and complexity of the generated application.
Time taken to complete this phase in milliseconds.
Measures the duration from phase initiation to completion, including all AI interactions, validations, and corrections. This metric helps identify performance characteristics of individual phases.
OptionalreasonOptional reason for failure, classified from the captured error or emitted by the harness.
"rate_limit": vendor API refused the request (OpenRouter 403/429 or
"Key limit exceeded" messages)."unknown": caught error that does not match a known classification."aborted_stagnation": the stagnation detector aborted a sub-loop after
observing a repeating failure signature."aborted_budget": the phase wall-time budget elapsed before the
sub-loop produced a result.Absent when the phase succeeded. The aborted_* and "rate_limit"
values are harness/infra outcomes and should be excluded from model-
quality scoring; "unknown" is a model-quality failure.
OptionalruntimeOptional runtime execution digest.
Populated only for the realize phase when the benchmark was run with runtime execution enabled. Summarizes the outcome of running the generated backend against its generated e2e test suite (see IAutoBeRealizeTestResult). Absent when runtime execution was not performed or the phase did not reach a runnable state.
Indicates whether the phase completed successfully.
True if the agent phase completed without errors and produced valid outputs that passed all validation checks. False indicates the phase encountered errors or failed validation, requiring investigation or retry.
State and metrics for an individual development phase.
Captures the execution results and performance characteristics of a specific agent phase in the vibe coding pipeline. This granular data enables detailed analysis of each phase's contribution to the overall development process and helps identify bottlenecks or areas for improvement.