postmortem

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a workflow for processing untrusted external data sources, specifically system logs, commit history, deployment data, and alerts. This creates a surface for indirect prompt injection where malicious content embedded in these sources could influence the generated report or content written to project documentation files. 1. Ingestion points: System logs, git commit messages, deployment logs, and alerts mentioned in the timeline construction workflow. 2. Boundary markers: No delimiters or boundary markers are defined to isolate the processed data from the agent instructions. 3. Capability inventory: The skill specifies updating project files including docs/harness/POSTMORTEM_TEMPLATE.md, RISKS.md, DECISIONS.md, and OPERATIONS_RUNBOOK.md. 4. Sanitization: No sanitization or validation steps are provided for the external data ingested during the workflow.
  • [NO_CODE]: The skill contains only markdown instructions and configuration metadata without any executable scripts or code, which limits the risk of direct execution of malicious logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 01:42 PM
Security Audit — agent-trust-hub — postmortem