The Agent Skills Directory

[COMMAND_EXECUTION]: The skill enables the execution of various show commands on Juniper network devices through the pyats_run_show_command tool. While documented as read-only, these commands provide deep access to system state and hardware status.
[CREDENTIALS_UNSAFE]: The skill retrieves configuration data for SNMP and NTP (show configuration snmp, show configuration system ntp). These configuration blocks frequently contain sensitive information, such as SNMP community strings or authentication keys, which are exposed to the agent's context.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted sources by reading system and firewall logs.
Ingestion points: show log messages and show firewall log in SKILL.md retrieve data generated by external network events.
Boundary markers: The instructions do not define boundary markers or delimiters to isolate log content from instructions.
Capability inventory: The skill uses pyats_run_show_command to interact with network infrastructure.
Sanitization: There is no evidence of sanitization or filtering of the log content before it is processed by the agent.

pyats-junos-system