internal-comms

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill is designed to ingest and summarize large volumes of untrusted data from internal platforms, creating a significant vulnerability surface (Category 8).\n
  • Ingestion points: Internal Slack channels, company-wide email threads, and Google Drive documents as specified in examples/3p-updates.md, examples/company-newsletter.md, and examples/faq-answers.md.\n
  • Boundary markers: Absent. There are no delimiters or 'ignore instructions' warnings used to wrap the ingested content, increasing the risk that the agent follows malicious instructions embedded in the data.\n
  • Capability inventory: The skill utilizes the agent's ability to read from organizational tools and generate textual reports; it does not include dangerous capabilities like shell access or file writing.\n
  • Sanitization: Absent. The skill lacks instructions to validate, escape, or sanitize content pulled from external sources before it is processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:49 PM