plugin-publishing

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow instructs adding and installing marketplaces from GitHub (e.g., "/plugin marketplace add /" and "/plugin install @") and states that installed plugins are scanned for SKILL.md files, meaning the agent will fetch and consume user-generated content from public GitHub repos which can change agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly directs users to register and install a marketplace by GitHub repo path (e.g., the runtime command "/plugin marketplace add Touricks/fanshi_personal_skills" / "/"), which causes the agent to fetch remote plugin/SKILL.md content from that repo that will directly control prompts, so the GitHub repo path (Touricks/fanshi_personal_skills / /) is a runtime external dependency that can control agent instructions.