kick-platform

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill includes multiple example requests and code that place client_secret and access tokens directly into request bodies/headers (e.g., client_secret=YOUR_CLIENT_SECRET and Authorization: Bearer <access_token>), which would require an LLM to handle or emit secret values verbatim if it produced executable requests—this creates an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s webhook handler at /webhooks/kick ingests outsider-authored chat/event payload fields (e.g., chat.message.sent -> content) from incoming HTTP webhook requests, which can carry attacker-controlled free text.