The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes a log analysis feature in scripts/agent.py that processes OT protocol logs provided by the user. As with any tool that ingests external data for AI-assisted review, there is a theoretical surface where malicious entries in the logs could attempt to influence the agent's interpretation of findings.
Ingestion points: The analyze_ot_protocols function in scripts/agent.py reads and parses external JSON lines logs.
Boundary markers: The skill does not currently define explicit boundary markers or 'ignore' instructions for the processed log data.
Capability inventory: The skill performs local file reading, report generation (writing JSON files), and standard HTTP GET requests to a configured API endpoint.
Sanitization: The script verifies JSON structure but does not implement specific content filtering for natural language instructions embedded within the logs.
[EXTERNAL_DOWNLOADS]: The skill correctly references and links to authoritative and well-known industry resources, including official documentation from Dragos, the International Society of Automation (IEC 62443), and CISA ICS Advisories.
[CREDENTIALS_UNSAFE]: The Python scripts and documentation use standard best practices for secret management, employing placeholders like your-api-key and <api_key> instead of hardcoding sensitive credentials.

implementing-dragos-platform-for-ot-monitoring