internal-comms

Pass

Audited by Gen Agent Trust Hub on Mar 28, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from multiple internal communication channels.
  • Ingestion points: Instructions in examples/3p-updates.md, examples/company-newsletter.md, and examples/faq-answers.md direct the agent to read content from Slack channels, Google Drive documents, Emails, and Calendar events.
  • Boundary markers: No boundary markers or specific delimiters are provided to the agent to distinguish between legitimate communication data and potentially malicious instructions embedded within those sources.
  • Capability inventory: The skill contains no executable code, but it has broad read access to organizational data silos. The resulting output is presented directly to the user.
  • Sanitization: No sanitization or validation logic is defined to filter or escape the content retrieved from external tools before it is processed into templates.
  • [NO_CODE]: The skill consists entirely of instructional Markdown files and does not contain any scripts, executables, or code-based logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 28, 2026, 08:17 PM
Security Audit — agent-trust-hub — internal-comms