text-gen-skill

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user prompts which are processed by an LLM, representing a surface for indirect prompt injection. However, the risk is negligible as the skill only returns text and lacks dangerous capabilities.
  • Ingestion points: prompt and systemPrompt parameters in apis/generateText.js and cloudfunctions/text-gen-handler/index.js.
  • Boundary markers: Absent. Input is appended directly to the messages array without delimiters.
  • Capability inventory: The skill is limited to generating text and returning it to the UI; it has no file system, command execution, or network exfiltration capabilities.
  • Sanitization: Absent. No filtering is performed on user input before processing.
  • [SAFE]: The skill leverages official platform APIs (wx.cloud.extend.AI) and verified vendor SDKs (@cloudbase/node-sdk, wx-server-sdk) from Tencent CloudBase. No unauthorized activities or suspicious domains were detected.
  • [SAFE]: The implementation includes a secure development 'preview mode' that allows testing with local static data, preventing unnecessary execution of cloud logic during the development phase.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 04:58 AM
Security Audit — agent-trust-hub — text-gen-skill