promptify

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

• Third-party content exposure detected (high risk: 0.80). The skill's web-researcher agent and the "+web" workflow (agents/web-researcher.md and commands/promptify.md, as noted in SKILL.md) explicitly perform WebSearch/WebFetch to retrieve and synthesize open web pages and documentation, which the agent reads and uses to shape prompts—creating a clear vector for untrusted third-party content to influence behavior.