ib-pitch-deck

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection through its core functionality of processing external data.
  • Ingestion points: The workflow involves extracting data from user-provided Excel, CSV, PDF, and Word documents, as well as databases and web sources (SKILL.md, Phase 1).
  • Boundary markers: The instructions lack specific guidance for the agent to use delimiters or to disregard potential instructions embedded within the processed source data.
  • Capability inventory: The skill instructs the agent to execute shell commands to convert and validate documents (SKILL.md, Phase 4).
  • Sanitization: There are no requirements or steps specified to validate or sanitize the external data before it is used to populate the presentation templates.
  • [COMMAND_EXECUTION]: The skill uses local system commands to verify the structural integrity of the generated presentations.
  • Evidence: The validation loop uses soffice --headless --convert-to pdf and pdftoppm to generate images from PowerPoint files for visual check (SKILL.md, Phase 4).
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 01:27 PM
Security Audit — agent-trust-hub — ib-pitch-deck