The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx chrome-use to run its core automation functions, which results in the dynamic download and execution of the chrome-use package from the npm registry. While npm is a standard service, fetching unversioned packages at runtime introduces a dependency on external code.
[PROMPT_INJECTION]: The skill's primary function is browser automation, creating an attack surface for indirect prompt injection where malicious instructions embedded in web pages could influence the agent's behavior.
Ingestion points: Untrusted data enters the agent context through the chrome-use open and chrome-use snapshot commands as described in SKILL.md.
Boundary markers: The instructions do not define delimiters or provide warnings to the agent to disregard instructions found within the processed web content.
Capability inventory: The skill grants the agent the ability to interact with web pages via click, fill, and other browser commands using the chrome-use CLI.
Sanitization: There is no evidence of content sanitization or validation for the data retrieved from external URLs before it is interpreted by the agent.

agentcore