decay

Pass

Audited by Gen Agent Trust Hub on Apr 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands, specifically mv, to relocate hypothesis files between project directories (e.g., from .fpf/knowledge/L2/ to .fpf/knowledge/L1/) as part of its governance workflow.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) due to how it processes external project data.
  • Ingestion points: Reads evidence files from .fpf/evidence/ and hypothesis files from .fpf/knowledge/L2/ (SKILL.md).
  • Boundary markers: Absent. The instructions do not define delimiters or provide guidance for the agent to ignore instructions that might be embedded within the files it reads.
  • Capability inventory: The skill has the ability to read and write files, as well as execute filesystem commands (mv) within the project structure (SKILL.md).
  • Sanitization: No validation or sanitization of data extracted from project files is performed before that data is used in report generation or shell command arguments.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 24, 2026, 05:13 AM