lazyweb-apply-design-best-practices

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

• Third-party content exposure detected (low risk: 0.10). The workflow fetches remote instruction files from GitHub via URL or Lazyweb MCP based on the user's design task matching a curated list of pointers, without ingesting monitored outsider-authored feeds.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to fetch external instruction files via raw GitHub URLs and third-party APIs during runtime to guide its behavior, which constitutes an operational dependency with unverifiable first-party provenance.

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

• Attempt to modify system services in skill instructions detected (medium risk: 0.40). The skill prompt instructs the agent to run curl -fsSL https://www.lazyweb.com/install.sh | bash if the Lazyweb MCP is missing, which executes an external shell script directly on the host machine.