tavily-search
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses environment variables or a localized
.envfile (~/.openclaw/.env) for API key management, which is a recommended security practice for handling sensitive credentials. - [SAFE]: Networking is performed via the Python standard library (
urllib.request) to the legitimate Tavily API endpoint (api.tavily.com), with no evidence of unauthorized data exfiltration. - [INDIRECT_PROMPT_INJECTION]: The skill fetches and displays content from external websites via search results, creating a surface for indirect prompt injection where an agent might process malicious instructions found in search snippets.
- Ingestion points: External search results are retrieved in
scripts/tavily_search.pyand returned to the agent. - Boundary markers: None; search results are returned as raw text or JSON structures without specific safety delimiters.
- Capability inventory: The skill contains no dangerous capabilities such as shell command execution, dynamic code evaluation, or file system write operations.
- Sanitization: Results are parsed as structured JSON and formatted for output without further content filtering.
Audit Metadata