meeting-briefing

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill is composed exclusively of Markdown instructions, methodology descriptions, and response templates. It does not include any Python scripts, Node.js code, shell scripts, or other executable components.
  • [SAFE]: Evaluation of the skill's instructions revealed no attempts at prompt injection, obfuscation, or unauthorized data access. The methodology described follows standard administrative and legal support practices.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a data aggregator, which creates a surface for indirect instructions to be processed from external sources.
  • Ingestion points: Untrusted data enters the agent's context through connected systems identified in Step 3, including Email, Chat (Slack/Teams), Documents (Box/Egnyte/SharePoint), and CRM/CLM systems.
  • Boundary markers: Absent. The skill does not instruct the agent to use specific delimiters or to disregard potential commands found within the processed materials.
  • Capability inventory: No capabilities such as file-writing, network requests, or command execution are requested or defined within this specific skill file.
  • Sanitization: There are no instructions for sanitizing or escaping the content retrieved from external sources before it is synthesized into the briefing template.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 12:08 PM
Security Audit — agent-trust-hub — meeting-briefing