openteam-control

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill operates by executing the openteamcli tool on the local machine to manage AI group chats, roles, and tasks.
  • [EXTERNAL_DOWNLOADS]: Recommends the installation of the @openteam/cli package from the official NPM registry to enable the skill's functionality.
  • [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface through the processing of task.json and roles.json files which define AI behaviors and tasks.
  • Ingestion points: Reads task and role configurations from local JSON files.
  • Boundary markers: None (uses structured JSON).
  • Capability inventory: Execution of all openteamcli subcommands (daemon, chat, role, task).
  • Sanitization: Relies on the user to ensure prompts do not contain sensitive data as per the included instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 11:48 AM
Security Audit — agent-trust-hub — openteam-control