The Agent Skills Directory

[SAFE]: The skill consists entirely of high-level natural language instructions and does not include scripts, executables, or external package dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill processes activity timelines from external hardware devices via the unifi_location_timeline tool. Ingestion points: Output from the unifi_location_timeline call. Boundary markers: None specified. Capability inventory: Restricted to interpreting events and presenting a narrative; no file-system, network, or command execution tools are authorized. Sanitization: None specified. The risk is considered negligible given the lack of exploitable capabilities.
[NO_CODE]: No code or scripts were provided with the skill, limiting the potential for malicious execution or obfuscation.

security-patrol