kuroco-api-content

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt includes code examples that hard-code passwords and static tokens (e.g., 'password123', 'your-static-token-here') and shows embedding tokens directly into request headers/bodies, which encourages the agent to place secret values verbatim into generated code or commands.

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

• Third-party content exposure detected (high risk: 0.70). The skill's SKILL.md includes runtime fetch calls to external Kuroco sites (e.g., https://{サイトキー}.g.kuroco.app/rcms-api/... in functions like fetchNewsList, fetchNewsDetail and fetchAllNews) which ingest user-generated CMS content (contents HTML/JSON) that the agent is expected to read/parse and which could influence subsequent API actions (e.g., updates/bulk_upsert), exposing it to untrusted third‑party content that could carry indirect prompt injections.