internal-comms
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and summarize large volumes of untrusted data from corporate tools like Slack, Email, and Google Drive.
- Ingestion points: The
examples/3p-updates.md,examples/company-newsletter.md, andexamples/faq-answers.mdfiles explicitly instruct the agent to gather context from Slack posts, Google Drive documents, and company-wide emails. - Boundary markers: None. The instructions do not include delimiters or warnings to treat ingested content as data rather than instructions.
- Capability inventory: The skill has the capability to read external data and synthesize it into executive-level communications, which could be used to leak sensitive information or misinform leadership if the agent follows instructions hidden within a Slack message or document.
- Sanitization: There is no mention of sanitizing or filtering the content retrieved from these external tools before it is processed by the LLM.
Audit Metadata