axolotl

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes a runtime function that fetches and parses arbitrary HTTPS configuration URLs (see cli.config.check_remote_config at https://docs.axolotl.ai/docs/api/cli.config.html), so remote content loaded at runtime can directly control prompts/configuration and thus represents a high-confidence external dependency risk.