pitch-deck-reviewer
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of processing untrusted external documents and web content. Adversarial instructions hidden within pitch decks could potentially influence the agent's behavior.
- Ingestion points: The workflow involves reading content from local PDF and PPTX files, as well as fetching data from external URLs via the
WebFetchtool (SKILL.md). - Boundary markers: The skill instructions do not define explicit delimiters or 'ignore instructions' wrappers to isolate untrusted deck content from system logic.
- Capability inventory: The agent is granted access to potentially powerful tools including
Bash,Write,Edit, and network-enabled tools likeWebFetchandWebSearch(SKILL.md). - Sanitization: No input validation or sanitization of the deck content is specified prior to analysis.
Audit Metadata