pitch-deck-reviewer
Pass
Audited by Gen Agent Trust Hub on Apr 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of analyzing untrusted external data using a broad toolset.
- Ingestion points: The skill reads user-provided files (PDF, PPTX, PNG, JPG) and fetches remote content from Google Slides links using 'WebFetch'.
- Boundary markers: Absent. The instructions do not specify any delimiters or directives to the agent to treat slide content as untrusted or to ignore instructions embedded within the pitch decks.
- Capability inventory: The skill has access to 'Bash', 'Write', 'Edit', 'WebSearch', and 'WebFetch', providing a significant attack surface if a malicious deck successfully overrides the system prompt.
- Sanitization: Absent. There is no mention of validating, filtering, or escaping the text extracted from the presentation files before processing.
Audit Metadata