vike-web-search
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external content from web search snippets, which is a potential surface for indirect prompt injection if the search results contain malicious instructions.
- Ingestion points: Results returned by the
vike web searchcommand as defined in SKILL.md. - Boundary markers: Search results are returned in a structured JSON format, helping the agent distinguish between data and instructions.
- Capability inventory: The skill restricts itself to the
vike:*namespace tools using theallowed-toolsfield, minimizing the risk of unauthorized command execution. - Sanitization: Relies on the JSON structure for data separation; no explicit text-level sanitization is defined in the instructions.
Audit Metadata