browser-devtools-cli

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADS
Full Analysis
  • Persistence (HIGH): The documentation instructs users to modify shell profiles like ~/.bashrc or ~/.zshrc with eval \"$(browser-devtools-cli completion bash)\". This allows the tool to execute code automatically at the start of every shell session, creating a persistence vector.
  • Indirect Prompt Injection (LOW): The skill is designed to process untrusted content from the web, which may contain malicious instructions. Evidence Chain: 1. Ingestion points: navigation go-to and content get-as-text. 2. Boundary markers: Absent; no instructions are provided to the agent to distinguish between tool output and page content. 3. Capability inventory: The skill uses Bash to execute commands, allowing for network access and file system interaction. 4. Sanitization: No sanitization or escaping of web content is performed.
  • Credentials Unsafe (LOW): Example code in the skill documentation contains a hardcoded password string (password123).
  • External Downloads (MEDIUM): The skill requires installing an unverified package from the public npm registry (browser-devtools-mcp).
  • Command Execution (LOW): The skill utilizes the Bash tool to run its core CLI functions, giving it extensive control over the browser environment and local state.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:25 PM