The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content retrieved from RHDH instances through playwright-cli snapshot. This content (e.g., entity descriptions, tab names, page text) could contain malicious instructions intended to influence the agent's behavior.
Ingestion points: Web page snapshots and text content retrieved via playwright-cli snapshot in SKILL.md.
Boundary markers: No specific delimiters or instructions are used to prevent the agent from interpreting embedded text as instructions.
Capability inventory: The skill uses the Bash tool to execute playwright-cli for browser interaction, navigation, and taking screenshots.
Sanitization: No sanitization or validation of the retrieved web content is performed before the agent processes it.

rhdh-playwright