market-research-digest
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it ingests untrusted data from multiple external sources. 1. Ingestion points: Files and links within the Google Drive research folder (INPUT_FOLDER) and Granola meeting notes from the last 30 days. 2. Boundary markers: The prompt template lacks delimiters or instructions to ignore embedded instructions in the source files. 3. Capability inventory: The skill can read private files, access meeting transcripts, search the web, and write files to Google Drive. 4. Sanitization: No evidence of sanitization or content validation for retrieved data.
Audit Metadata