gologin-web-access-skill

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill is implemented as a wrapper for the gologin-web-access CLI tool, which is used to perform all web scraping and browser automation operations. This is the intended and documented architectural design.
  • [EXTERNAL_DOWNLOADS]: The skill documentation recommends installing or running the gologin-web-access CLI via npm or npx. These resources are official vendor tools provided by the skill's author (gologinlabs) and are standard for this integration.
  • [DATA_EXPOSURE]: Tools such as browser_cookies, browser_storage_export, and browser_screenshot allow the agent to retrieve sensitive session information, which is a core requirement for browser automation. Additionally, parse_document allows for reading local files for analysis.
  • [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by processing untrusted data from external websites. Ingestion points: read_page, scrape_url, scrape_markdown, scrape_text, scrape_json, batch_scrape, search_web, map_site, crawl_site, and parse_document. Boundary markers: None identified in the prompt templates. Capability inventory: browser_eval (JavaScript execution), browser_upload (file upload), browser_cookies, browser_storage_export, and parse_document (local file access). Sanitization: None explicitly implemented in the skill instructions.
  • [REMOTE_CODE_EXECUTION]: The browser_eval tool enables the execution of arbitrary JavaScript within the browser context, and the workflow_run tool executes multi-step JSON-based runbooks. These features are provided for advanced web interaction and are consistent with the tool's primary purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 04:18 PM