synthetic-search
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted web search results, creating a surface for indirect prompt injection.\n
- Ingestion points: Web search result records are ingested from the Synthetic API via
scripts/search.jsandscripts/probe_synthetic_search.py.\n - Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between its commands and the search result content.\n
- Capability inventory: The skill allows network communication with
api.synthetic.newand returns content snippets to the caller.\n - Sanitization: The Node.js helper script performs basic sanitization by normalizing whitespace and truncating snippets to 300 characters.
Audit Metadata