opencode-command-creator
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the creation of configuration files that execute arbitrary shell commands using the platform's '!
command' syntax. This is the intended functionality for creating automated workflows (e.g., running tests or checking git logs). - [COMMAND_EXECUTION]: The 'scripts/create-command.py' script is used to generate command files on the local filesystem. It determines the file path based on the command name provided by the user, which presents a potential path traversal risk if used with untrusted input.
- [PROMPT_INJECTION]: The skill defines templates that interpolate user-supplied arguments directly into LLM prompts, creating a surface for indirect prompt injection. Ingestion point: $ARGUMENTS and positional parameters in command files. Boundary markers: Absent in provided templates. Capability inventory: Shell execution via '!
command' and file content reading via '@filename'. Sanitization: Not performed on input parameters.
Audit Metadata