unifi-network-setup
Warn
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill instructs the agent to ask the user for their UniFi controller username and password, which are sensitive administrative credentials.- [COMMAND_EXECUTION]: The skill uses "powershell -ExecutionPolicy Bypass" to run local scripts. This bypasses security configurations intended to prevent the execution of potentially malicious or unsigned scripts on Windows systems.- [COMMAND_EXECUTION]: Sensitive credentials (username, password, and API key) are passed as plaintext arguments in shell commands during the configuration write step. Command-line arguments are generally visible to other users on the system and are often logged by the operating system, creating a significant risk of local credential exposure.- [DATA_EXFILTRATION]: Use of plaintext environment variable assignments like "UNIFI_NETWORK_PASSWORD=" in shell commands can lead to secrets being stored in shell history files (e.g., .bash_history) or appearing in system-wide process monitors.- [COMMAND_EXECUTION]: The skill exposes a command injection surface because it interpolates user-provided data directly into shell commands without instructions for sanitization or escaping.
- Ingestion points: User input for "host", "username", "password", and "API key" entered via the AskUserQuestion tool in SKILL.md.
- Boundary markers: None. The user input is placed directly into shell commands.
- Capability inventory: The skill uses the Bash tool to execute set-env.sh and related scripts.
- Sanitization: There are no instructions for the agent to validate or escape characters (like semicolons or backticks) in the user input before execution, which could be exploited by a malicious user to execute arbitrary commands.
Audit Metadata