Skills
Navigate
skills/uncovertechtalent/cto-powersuit/incident/Gen Agent Trust Hub

incident

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements an 'overlay' feature that reads instructions from a path defined by the $POWERSUIT_OVERLAY environment variable.
  • Ingestion points: SKILL.md (Overlay section) specifies reading README.md and other files within the $POWERSUIT_OVERLAY/reliability directory.
  • Boundary markers: Absent; the skill lacks instructions to ignore potentially malicious instructions embedded in the overlay content.
  • Capability inventory: SKILL.md (Steps 1 and 3) utilizes shell commands (mkdir, rsync) and script execution (python3).
  • Sanitization: Absent; content from the overlay is trusted and followed without validation or filtering.
  • [COMMAND_EXECUTION]: The skill uses shell commands to initialize incident directories and generate status page artifacts.
  • Evidence: execution of mkdir, rsync, and python3 scripts/build_status.py in the Steps and Overlay sections of SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 12:51 PM