backend-setup-wizard

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to ask the user to provide an API key/token via chat and to immediately write that secret into a .env file, which requires the LLM to receive and handle secret values directly (and risks the model outputting or embedding them verbatim).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to web-search for and "fetch the actual page" of official provider documentation and then extract and follow those steps, meaning externally-hosted docs (provider docs URLs) would directly control the agent's prompts/instructions during execution.