internal-comms

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its reliance on ingesting data from external and internal sources that may contain malicious instructions.
  • Ingestion points: The skill explicitly instructs the agent to gather content from Slack messages, emails, Google Drive documents, Calendar events, and external press (as seen in 3p-updates.md, company-newsletter.md, and faq-answers.md).
  • Boundary markers: No delimiters or safety instructions are provided to help the agent differentiate between its core instructions and instructions that may be embedded in the data it processes.
  • Capability inventory: The agent uses its capabilities to search through and summarize internal company communications and documentation to produce newsletters and FAQs.
  • Sanitization: The skill lacks any steps for sanitizing or validating ingested content to prevent the execution of malicious prompts embedded within the messages or files it retrieves.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 06:46 PM