compete

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill utilizes WebSearch and WebFetch tools to ingest untrusted data from external websites, reviews, and job postings for competitive analysis. This creates a potential surface for indirect prompt injection. The skill mitigates this risk by requiring the agent to perform prompt-injection checks on all fetched content using a standardized safety protocol (_common/WEB_FETCH_SAFETY.md) and mandates that all findings must be backed by cited sources to prevent the unverified adoption of external instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 06:56 AM
Security Audit — agent-trust-hub — compete