conference-networking-prep

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses WebSearch and WebFetch tools to retrieve content from external conference websites, speaker lists, and search engine results for research purposes.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from external sources (Step 2 and Step 5) and uses it to generate icebreakers and networking playbooks. Instructions hidden in speaker bios or conference pages could theoretically influence the agent's behavior.
  • Ingestion points: Data retrieved via WebFetch from conference speaker/sponsor pages and WebSearch results for individual research.
  • Boundary markers: Absent; there are no instructions to the agent to ignore or delimit embedded instructions within the fetched content.
  • Capability inventory: WebSearch, WebFetch, and multiple Amplemarket MCP tools (enrich_company, search_people, enrich_person, create_lead_list).
  • Sanitization: Absent; the instructions do not specify any filtering or validation of the retrieved web content before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 11:36 PM
Security Audit — agent-trust-hub — conference-networking-prep