running-the-proxy

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill contains literal bearer tokens and example keys embedded in commands/config (e.g., the curl Authorization header, a virtual key in the TOML, and a Docker env example), which instructs the agent to include secret-like values verbatim and therefore risks secret exfiltration.

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

• Potentially malicious external URL detected (high risk: 0.90). The Docker instruction pulls and runs the external container image ghcr.io/xberg-io/liter-llm at runtime, which fetches and executes remote code (the image) required to run the proxy.