tiangong-kb-course-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs the wrapper/agent to forward api_key/bearer_token as CLI flags (e.g., --bearer-token) and emit Authorization: Bearer ... for course requests, which requires embedding secret values verbatim in generated commands/requests.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes "npx @tiangong-ai/cli@latest" at runtime (which fetches and executes remote code from the npm registry, e.g. registry.npmjs.org/@tiangong-ai/cli), and the fetched CLI is required to perform the search and thus can execute code or control agent behavior.