agent-browser

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and interact with untrusted external data from web pages, which serves as a vector for indirect prompt injection. 1. Ingestion points: Data enters the agent's context through agent-browser snapshot and agent-browser get text as described in SKILL.md. 2. Boundary markers: The documentation does not specify the use of delimiters or 'ignore' instructions for the content retrieved from websites. 3. Capability inventory: The skill possesses significant capabilities including network navigation (open), page interaction (click, fill), file system operations (state save, screenshot, pdf), and the ability to run local shell scripts (templates/). 4. Sanitization: No sanitization or content filtering mechanisms are mentioned for the data extracted from the browser.
  • [COMMAND_EXECUTION]: The skill documentation includes and references shell script templates (e.g., templates/form-automation.sh) used to automate browser workflows, which involves executing local script files.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of Appium via NPM to support mobile browser automation. This is a standard dependency for the tool's mobile testing functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:42 PM