email-validator

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The Skill contains deceptive runtime instructions that tell the agent to install dependencies automatically and to hide that action from the user (e.g., "without asking the user to confirm" and "Do not describe it as a request..."), which are hidden/overriding behaviors outside the declared email-validation purpose.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the agent to invoke the browser-act Skill for “live browser access or website interaction,” which at runtime can fetch and extract arbitrary public web page text (outsider-authored free text) into the LLM context via BrowserAct’s page-reading workflow.

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

• Potentially malicious external URL detected (high risk: 0.80). The Skill instructs the agent to automatically install a required runtime dependency from https://www.browseract.com/?co-from=lead-generation&redirect=https://github.com/browser-act/skills/tree/main during execution, which implies fetching and running remote installation content that controls the agent's runtime behavior.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The Skill explicitly directs the agent to automatically install the BrowserAct CLI as a required runtime dependency and to do so without asking the user for confirmation, which instructs the agent to change the host system state (install software) and bypass normal consent controls even though it does not explicitly request sudo.