ai-model-nodejs
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill uses official vendor resources from 'tencentcloudbase', including the '@cloudbase/node-sdk' package and official Tencent Cloud domains for documentation and resource purchasing. The inclusion of the MIT license and standard installation procedures aligns with legitimate development tools.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an inherent attack surface for indirect prompt injection as it processes external messages via AI models and performs configuration tasks (like 'UpdateAIModel') based on user input. Ingestion points: The 'messages' payload in 'generateText' and 'streamText' methods within 'SKILL.md' and 'references/api-reference.md'. Boundary markers: The skill lacks explicit prompt boundary markers but enforces a rigid two-step preflight process to verify environment and model state before execution. Capability inventory: Includes the ability to invoke cloud management APIs ('callCloudApi') and generate backend Node.js code. Sanitization: No specific sanitization or escaping of external content is documented for the prompt payloads, which is typical for SDK wrappers.
- [CREDENTIALS_UNSAFE]: The skill documentation demonstrates SDK initialization using 'secretId' and 'secretKey'. It correctly uses non-sensitive placeholders for these values and provides clear best-practice guidance, explicitly advising against hard-coding third-party API keys and recommending managed secret storage.
Audit Metadata