sendblue-cli

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill instructs the user to install the @sendblue/cli package via npm. This is a standard global installation of the vendor's own CLI tool, which is necessary for the skill's stated purpose.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill documents the local storage of credentials in ~/.sendblue/credentials.json with restricted file permissions (600). This is a recommended security best practice for CLI tools to avoid exposing secrets in environment variables or command history.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for the agent to send user-supplied strings as messages, creating a data ingestion surface.
  • Ingestion points: Phone numbers and message content are passed as shell arguments to the sendblue send command.
  • Boundary markers: None explicitly defined in the skill markdown body.
  • Capability inventory: The skill uses shell execution to interact with the system and network through the sendblue binary.
  • Sanitization: No specific sanitization or escaping logic is described in the instructions, though the CLI itself is the primary consumer of the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 02:26 AM
Security Audit — agent-trust-hub — sendblue-cli