miniprogram-development
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the use of
npxto load and execute the@cloudbase/cloudbase-mcppackage from the npm registry. This is an official vendor resource provided by Tencent CloudBase for Model Context Protocol integration. - [REMOTE_CODE_EXECUTION]: The skill utilizes the
npxcommand to dynamically download and execute the latest version of the CloudBase MCP toolset, which constitutes remote code execution from a package registry. - [COMMAND_EXECUTION]: The skill utilizes several CLI tools for development workflows, including
miniprogram-cifor project preview and upload automation, andmcporterfor managing CloudBase MCP server tools and interactive authentication. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its interaction with user-supplied project files and configurations.
- Ingestion points: The agent reads and processes
project.config.json, mini program page files (js, wxml, json), and other project-specific source code. - Boundary markers: The skill does not define explicit delimiters or instructions to the agent to disregard instructions embedded within the source files or configurations it processes.
- Capability inventory: The agent has the capability to write and modify project files on the file system and execute shell-based tools (mcporter, miniprogram-ci).
- Sanitization: There is no evidence of content sanitization or validation of the ingested project data before it is incorporated into the agent's context or used to drive tool execution.
Audit Metadata