internal-comms
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and summarize large volumes of untrusted data from Slack, Email, and Google Drive. If an attacker posts a malicious instruction in a high-traffic Slack channel or a shared document, the agent might interpret that data as a command while generating a newsletter or FAQ.
- Ingestion points:
examples/3p-updates.md,examples/company-newsletter.md, andexamples/faq-answers.mdall instruct the agent to crawl Slack channels, emails, and shared documents. - Boundary markers: There are no explicit instructions or delimiters (like XML tags) provided to tell the agent to treat the retrieved content purely as data and not as instructions.
- Capability inventory: The skill is primarily focused on text generation and summarization based on the provided instructions, but it requires broad read access to corporate communication tools.
- Sanitization: No sanitization or validation steps are defined for the ingested content before it is processed by the LLM.
Audit Metadata