lazycat-movie-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). Insecure: the prompt instructs embedding usernames/passwords directly into generated shell commands (e.g., --username '' --password '') and even includes default credentials (admin/adminadmin), which forces the LLM to output secret values verbatim.

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 and scripts (./scripts/yts-api.sh which fetches the public YTS API at https://movies-api.accel.li/api/v2 and can be pointed at arbitrary URLs, and the skill's SKILL.md) explicitly ingest and parse open-web movie/torrent data (including magnet/.torrent URLs) and then act on those results (adding selected magnets via ./scripts/vuetorrent-api.sh), so untrusted third‑party content can directly influence tool actions and decisions.

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

• Potentially malicious external URL detected (high risk: 0.70). The included runtime script ./scripts/yts-api.sh (and its copies) performs curl requests to the external API https://movies-api.accel.li/api/v2 and the returned JSON (torrents/magnet links) is injected into the agent workflow and used to drive prompts and actions, so this external URL directly controls agent behavior at runtime.