lazyweb-search-screens

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs the agent to accept and pass session IDs and to "give the user the private url" (i.e., output private resource URLs) and to include an agentic_search_id in subsequent calls, which requires emitting secret-like identifiers verbatim into requests or user-visible output.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). This is suspicious because it instructs piping a remote shell script (https://www.lazyweb.com/install.sh) directly to bash from an unverified domain — a common malware distribution pattern.

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

• Third-party content exposure detected (low risk: 0.10). [Exclusion] SKILL.md describes using Lazyweb MCP to run lazyweb_search_screens and finalize with selected references, but it does not specify any runtime ingestion path for outsider-authored free text beyond user-initiated search queries/results content that the workflow actively fetches only via selected references.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill instructs the operator to run a curl | bash installer which fetches and executes remote code from https://www.lazyweb.com/install.sh during runtime, creating a high-risk external execution dependency.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs running an external install script via curl ... | bash (SKILL.md:3-5), which asks the agent/user to execute a remote script that can modify the host system and bypass security, so it poses a high risk of compromising machine state.