The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/agent.py performs network connectivity checks using the Python socket library. It attempts to connect to specific OT-related ports (e.g., Modbus/502, S7comm/102, DNP3/20000) on target hosts provided by the user via command-line arguments. This activity is consistent with the skill's purpose of assessing industrial network security.
[INDIRECT_PROMPT_INJECTION]: The assessment script processes external data which creates a surface for potential injection if the resulting reports are later processed by an LLM without verification:
Ingestion points: Target hostnames/IPs via the --targets argument and control implementation status via the --controls-data JSON file in scripts/agent.py.
Boundary markers: The script does not utilize boundary markers or explicit instructions to ignore potentially malicious embedded content within the input files.
Capability inventory: The script is capable of performing network port scans (socket.connect_ex) and writing JSON report files to the local file system (json.dump).
Sanitization: The script does not perform validation or sanitization on the input strings or JSON content before incorporating them into the final report.

implementing-conduit-security-for-ot-remote-access