The Agent Skills Directory

[COMMAND_EXECUTION]: The references/api-reference.md file contains template curl commands for retrieving data from Tenable and Qualys APIs. These examples correctly use placeholders for sensitive credentials and target well-known security service domains.\n- [EXTERNAL_DOWNLOADS]: The skill links to official documentation and industry standards from trusted sources such as NIST, CISA, and reputable security vendors (Tenable, Nucleus Security). These links are informational and do not involve automated script execution or downloads from untrusted sites.\n- [METADATA_POISONING]: A minor inconsistency exists between the author name in the frontmatter ('mahipal') and the copyright notice in the license ('mukul975'). This is evaluated as a documentation discrepancy and does not present a security risk.\n- [INDIRECT_PROMPT_INJECTION]: The scripts/process.py script ingests data from CSV files. While it lacks explicit boundary markers for all input fields, the processed data is used solely for local reporting and mathematical calculations, presenting no significant vulnerability surface.\n- [SAFE]: The Python scripts (scripts/agent.py and scripts/process.py) process local vulnerability data to calculate security metrics and generate reports. The code was analyzed for dangerous operations such as network exfiltration, arbitrary command execution, and persistence mechanisms, and no such patterns were found.

building-vulnerability-aging-and-sla-tracking