project

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the orchestrator to ask the user to type secret values in-chat and then write them into .claude/settings.local.json (and potentially print the value as a fallback), which requires the LLM to receive and embed secret values verbatim in its output.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill configures and relies on external MCP servers and packages that will be fetched/executed at runtime (notably the context7 MCP via npx and an external GitHub MCP endpoint), e.g. "https://api.githubcopilot.com/mcp" and the plugin git URL "https://github.com/ravindra-gadekar/fullstack-dev-plugin.git", which indicates remote code/endpoints are required at runtime.