lazyweb-search-experiments

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt explicitly instructs the agent to "give the user the private url", which requires outputting a private URL (likely containing a secret token) verbatim and thus risks secret exfiltration.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). The skill instructs running a remote shell script (curl ... | bash) from https://www.lazyweb.com/install.sh — a direct .sh download and pipe-to-shell from an unverified domain is a high-risk distribution vector for malware.

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

• Third-party content exposure detected (low risk: 0.10). [The required runtime workflow calls a first-party “lazyweb_search_experiments” and then “lazyweb_agentic_search_finalize,” but the provided SKILL.md does not specify any reading of outsider-authored free text (e.g., user-generated or third-party submitted content via email/chat/issue tickets) beyond generic search results, so outsider prompt injection cannot be determined from this spec alone.]

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

• Potentially malicious external URL detected (high risk: 1.00). The skill includes a runtime instruction to run curl -fsSL https://www.lazyweb.com/install.sh | bash, which fetches and executes remote code from https://www.lazyweb.com/install.sh and therefore directly executes external code at runtime.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill explicitly instructs running a remote installer via curl ... | bash, which directs the agent to execute a downloaded script that can modify system files and obtain elevated privileges, so it pushes the agent to compromise host state.