360 Secure SkillHub
360 安全技能中心

Analyze Json Data

核心功能为分析JSON数据,生成结构化API设计文档或OpenAPI规范。提供标准、verbose两种执行模式,执行流程含输入验证、处理、输出校验等步骤,确保可靠可审计。典型场景是开发人员需基于JSON快速生成API规范。输入为JSON数据,输出为目标文档,附带执行摘要,错误时返回可操作提示。

wangjipeng977
wangjipeng977
热度 16.10
最近更新 2026-05-27 11:29:59
分类:开发依赖:内部调用大模型应用场景:API 集成 / 数据处理 / 报告撰写

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add analyze-json-data
Metadata
name
analyze-json-data
description
Use when (1) Analyze JSON data and generate a structured API design document or OpenAPI specification.
license
MIT
metadata
{"version":"1.0","category":"design","author":"wangjipeng","sources":["https://github.com/MiniMax-AI/skills"]}

Core Position

This skill solves the specific engineering problem described in the trigger conditions. It focuses on reliable, auditable execution with clear error reporting.

Modes

/analyze-json-data

Default mode. Performs the core task with standard configuration.

/analyze-json-data --verbose

Verbose mode. Includes detailed logging and intermediate results.

Execution Steps

  1. Parse input — Validate and parse the input data; report any structural issues
  2. Validate — Confirm all required fields/parameters are present and well-formed
  3. Execute — Perform the transformation/action with validated input
  4. Verify — Check the output is well-formed and complete
  5. Report — Return structured output with a brief summary of what was done

Mandatory Rules

Do not

  • Do not make up facts or claim actions were taken that were not
  • Do not hardcode API keys — use os.getenv("API_KEY") instead
  • Do not store sensitive user data beyond the current session
  • Do not exceed token budget without warning the user first
  • Do not activate for off-topic requests — return a brief decline message

Do

  • Validate all inputs before acting
  • Handle errors gracefully with actionable error messages
  • Log actions taken for auditability
  • State explicitly when you are uncertain or data is insufficient

Quality Bar

Criterion Minimum Ideal
Task completion Core task completes without error Full task with all edge cases handled
Error reporting Clear error message on failure Actionable error with suggested fix
User communication Brief summary of what was done Detailed log with reasoning for decisions
Input validation Rejects malformed input Accepts valid input, explains rejection for invalid

A good output is one where all edge cases are handled gracefully with clear user communication.

Good vs. Bad Examples

Scenario Bad Good
Task unclear Attempts anyway, may be wrong Asks for clarification before proceeding
Error Technical error message "Cannot [action]: [reason] — suggest [alternative]"
Partial success Reports "Done" Reports "Done, but [partial] not completed — see details"
Edge case Silently skips or crashes Documents edge case and handles gracefully