post-mortem-author
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's instructions, templates, and reference materials are focused on documentation and incident management processes. No malicious patterns, such as command execution, data exfiltration, or obfuscation, were detected in any of the analyzed files.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted incident data, which could potentially contain biased framing or instructions to misrepresent facts (as demonstrated in the included evaluation tasks). However, the skill explicitly provides a 'Blameless review' framework and strict guidelines (Step 2) to ensure the agent rejects blame language and focuses on system-level causes, effectively mitigating the risk of the agent adopting incorrect framing from the input data.
Audit Metadata