The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Python scripts to perform local file operations. These scripts use standard libraries (json, argparse, pathlib) to read from and write to specific configuration files in the user's home directory (e.g., ~/.config/manus/mcp_config.json). No arbitrary shell command execution or unsafe subprocess calls were detected.
[DATA_EXFILTRATION]: No network operations or external data transmissions are present in the provided scripts. The tools are limited to local file system management of the Manus configuration.
[REMOTE_CODE_EXECUTION]: The scripts do not use dynamic execution functions like eval() or exec() on untrusted input, nor do they download or execute remote scripts.
[CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or tokens were found. The configuration format supports an auth field, but the scripts default this to null and do not attempt to harvest or leak credentials.
[PROMPT_INJECTION]: The skill instructions and metadata are focused on operational tasks and do not contain patterns intended to bypass safety filters or override agent instructions.

manus-mcp-configurator