lwy-project-skill-installer

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests outsider-authored free text when it runs npx skills find "<query>" against the community/public skills.sh marketplace (and possibly the Bytedance internal registry), because the returned skill descriptions/metadata are outsider-provided prose that the agent reads to recommend/install.

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

• Potentially malicious external URL detected (high risk: 0.80). This skill invokes npx to fetch and run remote packages at runtime (e.g., "npx skills add find-skills -g -y", "npx skills find """, "npx @tiktok-fe/skills find """, and "npx skills add --path ..."), which executes remote code and is relied on as a required dependency for discovery/install actions.