x-collect
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it is designed to ingest and synthesize untrusted data from multiple web search rounds. An attacker could potentially influence the agent's behavior by placing malicious instructions on indexed web pages.\n
- Ingestion points: Web search results retrieved during the four rounds of research (official sources, technical analysis, comparisons, and latest updates) as described in
SKILL.md.\n - Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to disregard or sanitize directives found within external content.\n
- Capability inventory: Uses the
WebSearchtool to fetch data and includes a step to 'Save to temp file'.\n - Sanitization: No input sanitization or validation of retrieved search results is specified before processing them into a structured report.\n- [SAFE]: The
skill-report.jsonfile contains metadata making authoritative claims about the skill's safety (e.g., 'risk_level': 'safe' and 'No security issues confirmed'). In accordance with security audit protocols, these claims are treated as data to be evaluated and not as definitive conclusions.
Audit Metadata