exa-web-search-free
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill issues live web searches and full-page crawls (web_search_exa, exa-full.crawling_exa) and pulls code/docs from public sites like GitHub and Stack Overflow (get_code_context_exa), so the agent ingests untrusted, user-generated public web content that could carry indirect prompt injection.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill configures and then invokes a remote MCP provider at https://mcp.exa.ai/mcp (and the variant https://mcp.exa.ai/mcp?tools=...) which is called at runtime via mcporter and therefore causes the remote server to execute tool logic and supply instructions that control the agent's behavior.
Audit Metadata