The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the stably verify command-line interface to perform automated application testing. This involves running a CLI tool that controls a web browser to validate application features.
[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the stably tool via npm install -g stably or to run it directly using npx stably. These are standard methods for utilizing Node.js packages from the official registry.
[INDIRECT_PROMPT_INJECTION]: The skill accepts a natural language description as an argument to define the expected behavior of the application being tested. This represents a vulnerability surface where data could influence agent actions.
Ingestion points: The description argument provided to the stably verify command in SKILL.md.
Boundary markers: None identified; the description is passed as a string literal.
Capability inventory: Command execution of the stably binary and subsequent automated browser actions (navigation, form interaction, clicking, and scrolling).
Sanitization: Not explicitly defined in the skill documentation; sanitization is assumed to be handled by the platform's backend processing.

stably-verify