The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the npx shiplight test command to run and verify tests. This is the expected and primary function of the tool for the Shiplight framework.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8) because it ingests untrusted application data (DOM, logs) and uses it to inform its triage and code-generation logic.
Ingestion points: Processes local .test.yaml files and application data retrieved via inspect_page, get_browser_console_logs, and get_browser_network_logs.
Boundary markers: No specific delimiters or instructions are defined to isolate untrusted application content from the agent's core logic.
Capability inventory: The skill is capable of modifying local YAML test files, interacting with the browser via act, and executing shell commands through the Shiplight runner.
Sanitization: No explicit sanitization or validation of external content is performed before the agent processes it to generate test fixes.

triage