pitch-deck

Warn

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (MEDIUM): The skill instructs the agent to execute shell commands using grep to extract content from local reference files.
  • [COMMAND_EXECUTION] (MEDIUM): The skill executes a Python script (scripts/create_pitch_deck.py) to process data and generate files. The source code for this script was not included in the provided files, making its behavior unverifiable.
  • [EXTERNAL_DOWNLOADS] (LOW): The skill requires the installation of the python-pptx package from an external repository (PyPI).
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill processes untrusted user input into a JSON structure (pitch_data.json) which is then consumed by a script.
  • Ingestion points: User-provided business details collected in Step 1 of the workflow.
  • Boundary markers: Data is structured into JSON keys; however, no specific delimiters or instructions exist to prevent the agent from obeying commands embedded within user data.
  • Capability inventory: The skill can execute shell commands (grep) and spawn Python processes.
  • Sanitization: No evidence of sanitization or validation of user-provided strings before they are written to the JSON data file.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 19, 2026, 03:51 PM