incident-postmortem-builder
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration attempts were detected. The skill follows its stated purpose and operates within a local environment.
- [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection by processing external timeline data from CSV files and interpolating it into Markdown reports.
- Ingestion points: The
--timelineCSV file read by thebuild.pyscript during runtime. - Boundary markers: Absent; the script does not implement delimiters to distinguish untrusted CSV content from the report template.
- Capability inventory: File system read (timeline data) and file system write (generated report output) via
build.py. - Sanitization: No sanitization or escaping is applied to fields like
event,actor, orsourcebefore they are embedded into the output Markdown table.
Audit Metadata