fact-check

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

• Third-party content exposure detected (low risk: 0.10). The skill workflow reads and fact-checks arbitrary web pages and primary sources fetched via headless checker subagents, representing low-risk indirect prompt injection from fetched web content.

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

• Potentially malicious external URL detected (high risk: 0.70). An inbound operational dependency exists because the skill instructs users to install external packages from npm (@zai-org/zcode-cli, @anthropic-ai/claude-code, @openai/codex) and fetch installer scripts from remote URLs (curl -fsSL https://opencode.ai/install | bash), which provide behavior-selecting configuration and execution code for the runtime environment under weak/unknown provenance (high).