create-tool
Audited by Socket on Feb 19, 2026
1 alert found:
Security[Skill Scanner] Backtick command substitution detected All findings: [HIGH] command_injection: Backtick command substitution detected (CI003) [AITech 9.1.4] [HIGH] command_injection: Backtick command substitution detected (CI003) [AITech 9.1.4] This skill is documentation and examples for creating Vapi tools. It does not contain embedded malicious code or obfuscation. The functionality requested (network access and a VAPI_API_KEY) is appropriate and proportionate for the stated purpose. The primary security considerations are operational: protect the VAPI_API_KEY, secure the developer's tool server, validate and sanitize tool call arguments, and ensure downstream integrations (Google, Slack, MCP) are configured with least privilege. Treat telephony-capable tools (transferCall, endCall, dtmf) as higher-risk capabilities and restrict who can create/update such tools. Overall the content looks BENIGN but operators should follow normal security hygiene when implementing the server and running provided commands. LLM verification: The provided SKILL.md is documentation and straightforward example code for implementing Vapi tools and a tool server. It contains no explicit malicious code or obfuscated payloads. The most significant risks are operational and design-related: powerful telephony actions and integrations with external services can be misused or misconfigured, potentially exposing PII or routing calls to unauthorized endpoints. The static scanner warnings are false positives (template literals). Recommend strict