The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a series of shell commands, including a local bootstrap script (scripts/vss-bootstrap.sh), git, curl, and jq. These are used to resolve the application root, perform health checks, and interact with the VSS API.
[EXTERNAL_DOWNLOADS]: The bootstrap script downloads source code from the author's GitHub repository (github.com/open-edge-platform/edge-ai-libraries.git) using git clone if a local checkout is not detected. This is a functional requirement for the skill to access the VSS application logic.
[PROMPT_INJECTION]: The skill processes user-supplied information, such as video titles and IDs, which are then used as arguments in shell commands (curl). This creates a surface for indirect prompt injection.
Ingestion points: Video title and videoId fields provided by users in prompts or resolved from prior tool outputs.
Boundary markers: The skill does not define specific delimiters or instructions for the agent to sanitize or ignore instructions embedded within these user-provided strings.
Capability inventory: The skill has the ability to execute shell commands (curl, bash) and perform network operations.
Sanitization: No explicit validation or escaping logic is provided for the interpolation of user strings into command arguments.

vss-summarize-video