The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands to invoke Python scripts (MCP tools) for network diagnostics. These commands utilize environment variables such as $MCP_CALL and $PYATS_MCP_SCRIPT to determine execution paths.
[DATA_EXPOSURE]: The troubleshooting procedures gather sensitive network infrastructure data, including routing tables, BGP/OSPF neighbor states, and access-lists. This is consistent with the skill's primary purpose of network analysis.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process output from external network devices and inventory systems (NetBox), which represents a potential injection surface.
Ingestion points: Data enters the agent context via output from pyats_run_show_command and netbox_get_objects calls documented in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the troubleshooting templates.
Capability inventory: The skill has the capability to execute network commands and record data to an audit trail via GAIT_MCP_SCRIPT.
Sanitization: No explicit sanitization or validation of the device output is described in the instruction text.

pyats-troubleshoot