The Agent Skills Directory

[SAFE]: The skill provides a legitimate set of instructions for web automation. All commands and examples are consistent with the stated purpose of browser testing and data extraction.
[PROMPT_INJECTION]: The skill's functionality involves processing data from external web pages, which presents a surface for indirect prompt injection. This is an architectural risk inherent to browser-based agents.
Ingestion points: Data enters the context through snapshot, get text, and find commands in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for content retrieved from the web.
Capability inventory: The skill includes interaction capabilities such as click, fill, and open in SKILL.md.
Sanitization: There is no evidence of sanitization or validation of external content prior to processing by the agent.

agent-browser