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

work-item

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' mechanism that directs the agent to read and follow instructions from a filesystem path defined by the $POWERSUIT_OVERLAY environment variable. This allows external data to override the skill's behavior.
  • Ingestion points: Files located in $POWERSUIT_OVERLAY/sdlc/, including README.md and content in sdlc/work-item/.
  • Boundary markers: No delimiters or specific 'ignore' instructions are provided to separate overlay content from core instructions.
  • Capability inventory: The skill utilizes GitHub CLI (gh), local Python execution (python3), and filesystem access.
  • Sanitization: No sanitization or validation of the content read from the overlay path is performed.
  • [COMMAND_EXECUTION]: The skill instructs the agent to run various commands to manage work items.
  • Evidence: Use of gh issue view to inspect tickets.
  • Evidence: Execution of python3 .github/actions/gate/run.py to perform gate checks.
  • Evidence: A shell check ls -1 "$POWERSUIT_OVERLAY/sdlc" to detect the presence of customization files.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 12:50 PM