agent-browser

Fail

Audited by Snyk on Mar 2, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 1.00). The prompt's examples and inputs explicitly show embedding credentials (e.g., password fields, proxy_username/proxy_password) directly into JSON/CLI commands, which would require an LLM to include secret values verbatim in its output.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly lets the agent open arbitrary URLs and perform web scraping/interactions (see the "open"/"goto" functions and examples that navigate to external pages like "https://google.com"), so the agent will fetch and interpret untrusted third‑party web content as part of its workflow.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 1.00). The Quick Start includes a command that pipes a remote installer to the shell (curl -fsSL https://cli.inference.sh | sh), which fetches and executes remote code and is presented as a required CLI installation for running the skill.
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 2, 2026, 09:24 AM