self-host-agenta
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the official documentation at docs.agenta.ai and the source code repository on GitHub. These links are used to guide the user through the installation process using official vendor resources.
- [COMMAND_EXECUTION]: The instructions involve executing local shell commands for deploying containers via docker compose, testing API health with curl, and generating secure keys using openssl. These operations are standard for the skill's purpose of managing a self-hosted software stack.
- [DATA_EXFILTRATION]: The skill contains a procedure for sending anonymous setup feedback to a public topic on ntfy.sh. The process includes specific safety instructions that require explicit user consent and the redaction of all secrets, API keys, and identifying data.
- [PRIVILEGE_ESCALATION]: In the troubleshooting guide, the user is advised how to grant themselves Docker access using sudo if they encounter permission errors. The documentation notes that this is a root-equivalent privilege on the host system.
Audit Metadata