md2wechat

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Discovery First" and "Agent Rules" explicitly instruct running md2wechat provider/prompt discovery commands (e.g., "md2wechat providers list/show --json", "md2wechat prompts show --json") and to prefer prompt metadata from those providers, and it also allows downloading arbitrary remote images (e.g., "md2wechat download_and_upload https://..."), so the agent fetches and interprets untrusted third-party content that can change which prompts/models/actions it selects.