sendblue-notify

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx @sendblue/cli, which fetches and executes the Sendblue command-line tool from the npm registry.\n- [COMMAND_EXECUTION]: Provides examples for executing shell commands to send notifications, including branching logic based on task success or failure.\n- [DATA_EXFILTRATION]: By design, this skill sends information (task status, branch names, etc.) to the Sendblue service. The instructions include explicit warnings against including sensitive data or secrets in notification messages.\n- [PERSISTENCE]: Recommends configuring a Stop hook in .claude/settings.json to automate notifications, which is a standard usage of the platform's lifecycle hooks to maintain behavior across sessions.\n- [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it interpolates command outputs (like $(deploy-status)) into notification messages. However, the instructions provide guidance on safe copy composition and require user confirmation, mitigating the risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 07:15 PM