hubspot

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION] (CRITICAL): The skill documentation explicitly instructs users to pipe a remote shell script to bash for installation: curl -sSL https://canifi.com/skills/hubspot/install.sh | bash. This is a classic RCE vector where the remote server can change the script content at any time to execute malicious commands.
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill relies on external scripts and tools from canifi.com, which is not a trusted source according to security guidelines. This includes both the primary skill installer and a prerequisite tool: curl -sSL https://canifi.com/install.sh | bash.
  • [COMMAND_EXECUTION] (HIGH): The setup process encourages the use of shell commands to install components and manage environment variables, bypassing standard package management and verification processes.
  • [CREDENTIALS_UNSAFE] (MEDIUM): The skill prompts users to store sensitive data including HUBSPOT_API_KEY, SERVICE_EMAIL, and SERVICE_PASSWORD. While it claims to store these locally via canifi-env, the insecure installation method of that tool makes the security of these credentials unverifiable.
Recommendations
  • HIGH: Downloads and executes remote code from: https://canifi.com/install.sh, https://canifi.com/skills/hubspot/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 17, 2026, 05:34 PM
Security Audit — agent-trust-hub — hubspot