The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes system commands using python3 to run the pyats_run_show_command function. This behavior is the primary intended function of the skill for network automation and troubleshooting.
[DATA_EXPOSURE]: The skill utilizes the PYATS_TESTBED_PATH environment variable to locate a testbed file. This file typically contains network topology information and device credentials required for connection. This is standard practice for pyATS-based automation.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data from network device CLI outputs (e.g., routing tables and neighbor states).
Ingestion points: CLI output from various show commands in SKILL.md.
Boundary markers: None present.
Capability inventory: Executes subprocesses via python3 as defined in SKILL.md.
Sanitization: No explicit sanitization of device output is performed before it is presented to the agent.

pyats-routing