taubyte-core-rules
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to automate Taubyte platform tasks by executing CLI commands for 'tau', 'dream', and 'docker'. This includes launching Docker Desktop on Windows via its standard file path and modifying the system 'hosts' file ('C:\Windows\System32\drivers\etc\hosts') for local DNS resolution. While these are privileged or sensitive operations, they are core to the skill's functionality and the instructions explicitly include a protocol to ask for user permission before modifying system files.
- [PROMPT_INJECTION]: The skill ingests user input to define resource names and configurations, which are then used in CLI commands and generated files. This creates a surface for indirect prompt injection as there are no specified sanitization or validation rules for this user-provided data.
- Ingestion points: User requests for creating websites, databases, or functions entering the context through natural language requests.
- Boundary markers: The instructions do not define delimiters or markers to separate user data from agent instructions in generated configurations or command arguments.
- Capability inventory: The skill can write configuration and code files, execute system commands ('tau', 'dream', 'git'), and perform network operations via the Taubyte CLI.
- Sanitization: No explicit validation or sanitization of user-provided inputs is mentioned before they are used in automation tasks.
Audit Metadata