The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes subprocess.run and asyncio.create_subprocess_exec to interact with system tools including git and the opencode binary. These executions are scoped to environment verification and protocol management.- [COMMAND_EXECUTION]: In CLI mode, the skill generates bash scripts from templates and executes them. The implementation mitigates shell injection risks by applying shlex.quote to all user-provided variables and validating environment variable keys.- [REMOTE_CODE_EXECUTION]: The skill uses python3 to execute a hardcoded script for log processing. This script includes a security check to prevent symlink-based file overwrite attacks, ensuring only intended logs are modified.- [DATA_EXFILTRATION]: The skill spawns a local HTTP server for Agent Client Protocol (ACP) communication. By default, it binds to 127.0.0.1, which restricts access to the local machine and establishes a standard local security boundary.

opencode-dispatch