The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses docker exec to run shell commands and Python scripts within the vss-agent container. This is used to discover environment variables such as HOST_IP and VLM_BASE_URL and to parse local YAML configuration files to determine model settings.\n- [EXTERNAL_DOWNLOADS]: Multiple curl commands are used to interact with local network services within the VSS stack, including VST (port 30888), VA-MCP (port 9901), and various VLM endpoints for health checks and data retrieval.\n- [PROMPT_INJECTION]: The skill processes untrusted incident data and VLM-generated descriptions which presents an indirect injection surface.\n
Ingestion points: Data is ingested from video_analytics__get_incidents (SKILL.md) and VLM completions.\n
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the report templates.\n
Capability inventory: The skill can execute container commands via docker exec and perform network requests using curl.\n
Sanitization: Employs jq for basic JSON parsing but does not sanitize the text content of incidents or captions.

vss-generate-video-report