playwright-cli-cloakbrowser

Fail

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the official Docker installation script from the well-known service at https://get.docker.com.
  • [REMOTE_CODE_EXECUTION]: Executes the Docker installation script by piping the output of a network request directly to the system shell (curl | sh).
  • [COMMAND_EXECUTION]: Utilizes playwright-cli, docker, and curl to perform automated browser actions and manage system containers.
  • [PROMPT_INJECTION]: The skill ingests and interprets data from external websites via accessibility snapshots and JavaScript evaluation, creating a surface for indirect prompt injection.
  • Ingestion points: External web content is ingested through the playwright-cli snapshot and eval commands.
  • Boundary markers: The instructions do not specify the use of delimiters or provide the agent with guidance to ignore instructions that might be embedded in the processed web data.
  • Capability inventory: The skill allows the agent to execute shell commands via playwright-cli, curl, and docker tools.
  • Sanitization: No evidence of sanitization or validation of the retrieved web content was found in the skill instructions.
Recommendations
  • HIGH: Downloads and executes remote code from: https://get.docker.com - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Aug 6, 2026, 03:56 PM
Security Audit — agent-trust-hub — playwright-cli-cloakbrowser