upskill

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). These URLs include direct raw shell-script downloads (curl|bash) from a GitHub repo and a reference to an external/unknown GitHub repo used for npx installs — executing remote .sh from unverified sources and installing code from unvetted repos is a high-risk distribution vector for malware.

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

• Third-party content exposure detected (high risk: 0.75). Outsider-authored free text can enter the LLM context via the runtime session artifacts that UpSkill captures (e.g., SessionEnd saves session.log and extracts [USER] messages), which are not authored by the operating user and are then fed to the Teacher/Student models during /upskill-build and later served via generated SKILL.md/CLAUDE.md.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's installer explicitly runs remote code at runtime (curl -sSL https://raw.githubusercontent.com/HKUDS/Upskill/main/cc-integration/install.sh | bash and/or git clone https://github.com/HKUDS/UpSkill && ... && bash install.sh), which fetches and executes upstream content required for installation.