web-search-plus
Fail
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: CRITICAL
Full Analysis
- [SAFE]: The skill implements proactive SSRF (Server-Side Request Forgery) protection in both the search and setup scripts. It validates hostnames and resolves IP addresses to block access to private networks, loopback addresses, and cloud metadata endpoints (e.g., 169.254.169.254) when connecting to self-hosted SearXNG instances.
- [SAFE]: Credential management follows security best practices. API keys are retrieved from environment variables or a local configuration file that is git-ignored by default, preventing sensitive tokens from being accidentally committed to version control.
- [SAFE]: The 'malicious URL' identified by scanners (searx.your-domain.com) is a non-functional documentation placeholder. It is used exclusively as a descriptive example in user-facing documentation and interactive setup prompts, posing no risk of automatic invocation or redirection.
- [SAFE]: The skill uses standard Python libraries for network requests and file operations. Analysis of the scripts confirms the absence of dangerous patterns such as obfuscation, dynamic code execution (eval/exec), or unauthorized privilege escalation.
- [SAFE]: The skill processes data from external search providers in a structured JSON format. This approach provides a clear boundary for the agent, reducing the risk of instructions embedded in search results being misinterpreted as core commands.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata