ic-onboarding

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to capture secret values (device_code, agent_token) and embed them verbatim into API calls, config files, and Authorization headers (e.g., writing ~/.config/ic/agent.json, curl -H "Authorization: Bearer $AGENT_TOKEN", or --header "Authorization: Bearer <agent_token>"), so the LLM must handle/output secrets directly.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs downloading and running a remote helper script (used as a headersHelper) from https://www.immersivecommons.com/tools/ic-mcp-auth/ (and also points to the inbox TUI installer at https://www.immersivecommons.com/tools/ic-inbox-tui/), which are runtime-fetched artifacts that the agent would execute/install, so this is a high-confidence runtime external dependency that executes remote code.