ai-model-web

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references documentation and sibling configurations from the vendor's official repository at cnb.cool (Tencent's cloud-native build platform). This is standard practice for maintaining up-to-date integration guides.
  • [COMMAND_EXECUTION]: Uses the callCloudApi tool to interact with the Tencent Cloud (tcb) service. This is used for legitimate administrative tasks such as verifying subscription status (DescribeEnvPostpayPackage), listing available models (DescribeAIModels), and enabling specific models (UpdateAIModel). These operations are restricted to the environment's authorized cloud identity.
  • [PROMPT_INJECTION]: As a model-invocation skill, it processes user-provided content through the messages array in generateText and streamText methods.
  • Ingestion points: messages parameter in SDK calls within SKILL.md.
  • Boundary markers: No explicit delimiters are suggested in the code snippets; however, the skill emphasizes structured API usage over raw string interpolation.
  • Capability inventory: The skill can modify environment configurations via UpdateAIModel and CreateAIModel using the callCloudApi tool.
  • Sanitization: The skill does not explicitly mention input sanitization, which is typical for low-level SDK integration skills but presents a standard surface for indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 04:24 PM
Security Audit — agent-trust-hub — ai-model-web