The Agent Skills Directory

[PROMPT_INJECTION]: No attempts to bypass safety filters or override system instructions were detected. The skill uses emphatic language to define legal guardrails for housing advertisements, which is appropriate for its stated purpose.
[EXTERNAL_DOWNLOADS]: The documentation includes a command to download the skill file from the author's GitHub repository (github.com/fourteenwm). This is a standard installation procedure from the vendor's official source.
[DATA_EXFILTRATION]: No network exfiltration or sensitive data access patterns were identified. The skill's configuration restricts access to the Read tool only, preventing network-based data transmission.
[COMMAND_EXECUTION]: No shell commands or executable script logic are present in the skill files. The instructions are purely textual guidelines for the AI agent.
[INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted campaign data, its purpose is to act as a safety filter. Given the restricted toolset (Read only) and the lack of dangerous capabilities, the attack surface is minimal.

fair-housing-compliance