The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines procedures for the agent to ingest and analyze firewall logs and intrusion detection alerts which contain data from external network sources.
Ingestion points: Logs such as /var/log/suricata/eve.json and CrowdSec decision lists (cscli decisions list) mentioned in SKILL.md and references/plugins.md.
Boundary markers: The instructions do not currently include explicit delimiters or prompts to ignore embedded instructions within processed log data.
Capability inventory: The skill possesses the capability to modify firewall rules (pfctl), manage system services (configctl), and interact with the OPNsense REST API (curl).
Sanitization: No specific sanitization or validation steps are outlined for the agent when interpreting the contents of these log files.
[EXTERNAL_DOWNLOADS]: The documentation includes references to external migration tools hosted on GitHub.
Evidence: The references/plugins.md file points to the meyergru/iscdhcp_to_dnsmasq and dreary-ennui/Convert-OPNSenseISCDHCPtoDNSMasqDHCP repositories for DHCP migration assistance.

opnsense-pfsense