gotify

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill permits and documents overriding tokens via a --token CLI argument and instructs the agent to run user-provided commands verbatim, which can force the LLM to emit user-supplied API tokens/credentials verbatim (high exfiltration risk).

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

• Third-party content exposure detected (medium risk: 0.30). At runtime the skill’s ws-subscribe command connects to the user-specified Gotify /stream WebSocket and ingests outsider-authored message JSON via the on_message callback (scripts/gotify_client.py: lines 175-205, especially 190-192).