core-agent-browser
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill documents the usage of a browser automation CLI tool. The commands described, such as navigation, snapshotting, and element interaction, are standard for web automation workflows and do not exhibit malicious patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest data from external websites via a browser, which introduces a surface for indirect prompt injection.
- Ingestion points: Data enters the agent context through
agent-browser snapshot,agent-browser get text,agent-browser get value, andagent-browser get titlecommands documented inSKILL.md. - Boundary markers: The skill instructions do not specify explicit delimiters or "ignore embedded instructions" warnings for the content retrieved from the browser.
- Capability inventory: The skill utilizes the
agent-browserCLI to perform navigation, interactive element manipulation (click, fill, type), and information retrieval from remote URLs. - Sanitization: No specific sanitization, filtering, or validation logic is defined within the instructions for handling the untrusted external content returned by the browser commands.
Audit Metadata