Outputs
Each successful or failed run can produce local artifacts and optional SQLite records.
Artifact Files
Runs are saved automatically under:
output/json/<timestamp>_<run_id>.json
output/txt/<timestamp>_<run_id>.txt
The JSON artifact contains structured parser, executor, and assessment data. The text artifact is a human-readable report.
Assessment Sections
The final assessment output can include:
| Section | Description |
|---|---|
buildspec_resolution_scope |
How the BuildSpec scope was resolved. |
root_cause_synthesis |
Cross-domain synthesis from logs, traces, and metrics. |
findings |
Structured findings reported by executor sub-agents. |
anomalies |
Anomalies detected during telemetry analysis. |
preliminary_causes |
Candidate causes inferred from the executor output. |
final_reporting |
Final report fields requested by the BuildSpec. |
SQLite Knowledge DB
When memory is enabled, Aware writes runtime information to SQLite.
Default URL:
sqlite:///output/assess.db
Tables:
runseventstask_resultsfindings
This database is useful for later inspection, debugging, and comparing assessment runs.