The Agent Skills Directory

[DATA_EXFILTRATION]: The skill handles sensitive BT-Panel API tokens which are required for authentication. These credentials are provided by the user and stored in a local YAML configuration file at ~/.openclaw/bt-skills.yaml. This is standard behavior for command-line tools that manage remote server infrastructure.
[COMMAND_EXECUTION]: The skill consists of Python scripts that perform network operations to interact with remote BT-Panel instances. The BtClient implementation in bt_common/bt_client.py explicitly disables SSL verification (verify=False) to support connections to panels using self-signed certificates, which is a common configuration in internal server environments.
[PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection due to its ability to read and display remote server logs.
Ingestion points: Remote log files (e.g., Nginx, Apache, MySQL, Redis) are retrieved and displayed to the agent via scripts/logs.py.
Boundary markers: The skill does not wrap log content in specific delimiters or provide explicit instructions to the agent to disregard instructions found within the logs.
Capability inventory: The skill has the ability to perform network requests and modify its local configuration file.
Sanitization: Log content is retrieved and formatted for display using the rich library but is not sanitized for potential embedded malicious instructions.

btpanel