event-staffing-ordering

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill performs network operations to external domains not on the predefined whitelist. It interacts with an MCP server at mcp.tempguru.co to fetch live data about cities, roles, and pricing. It also provides external links to tempguru.co for manual request submission.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided event data which acts as a potential attack surface for indirect prompt injection.
  • Ingestion points: User-provided event requirements such as city, dates, headcount, and special requirements (Workflow Step 1).
  • Boundary markers: Includes clear instructions for the agent to avoid gathering sensitive information like payment details, private attendee data, or credentials.
  • Capability inventory: The skill utilizes network tool calls to an external API and generates outgoing links for the user to visit.
  • Sanitization: There is no evidence of explicit input sanitization or schema validation for the data gathered from the user before it is passed to the MCP tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 01:19 AM