agent-browser
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill introduces a potential indirect prompt injection vulnerability surface because it is designed to browse, scrape, and interact with external, user-controlled or arbitrary web pages.
- Ingestion points: Web pages, forms, and external sites navigated during browser automation (documented in
SKILL.md). - Boundary markers: No specific delimiters or boundary rules are defined in this discovery stub to isolate untrusted web content from the agent's core instructions.
- Capability inventory: Navigating pages, filling forms, clicking buttons, extracting data, and automating desktop applications like Slack or Electron apps (documented in
SKILL.md). - Sanitization: No data sanitization or input validation mechanisms are described or enforced in this configuration file.
- [EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install a global package (
agent-browser) from the public npm registry to enable its capabilities.
Audit Metadata