302ai-api-integration

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the assistant to ask the user for their 302.AI API key and to embed that key verbatim into generated code and request headers (e.g., API_KEY = "{user_API_KEY}", "Authorization: Bearer ..."), which requires the LLM to handle and output secrets directly, creating a high exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to auto-fetch the public API list (via scripts.parse_api_list.fetch_llms_txt and references to https://doc.302.ai/llms.txt) and to use WebFetch to retrieve and parse user-selected API documentation pages (doc.302.ai links) which the agent must interpret to generate integration code, exposing it to untrusted third‑party content that could contain injected instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime explicitly auto-fetches and parses the external API list (https://doc.302.ai/llms.txt) via its parse script and uses WebFetch to load user-selected docs (e.g., https://doc.302.ai/147522039e0.md), and that remote content is parsed and injected into the agent's prompts/integration code generation, so it is a required runtime dependency that directly controls agent instructions.