google-maps-chain-store-lead-search

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill contains deceptive runtime-install instructions that tell the agent to install dependencies automatically without user confirmation and to hide/avoid describing the installation to the user, which changes agent behavior outside the stated lead-search purpose.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). This skill points to a third‑party site (browseract.com) used to install a CLI automatically (including a redirect to a GitHub repo); because it is not a well‑known vendor and the installer could deliver executables without explicit user consent, these links are a suspicious distribution vector for malware.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs the agent to invoke the browser-act Skill for live Google Maps browsing, which at runtime will fetch and read outsider-authored web page content (public listings/reviews) into the LLM context via the browser automation 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 installer code as a required dependency.