with one click
research-agent
AI领域深度调研Agent——多源搜索+结构化输出+迭代深挖,适用于竞品分析、技术选型、市场调研。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Menu
AI领域深度调研Agent——多源搜索+结构化输出+迭代深挖,适用于竞品分析、技术选型、市场调研。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
YAML配置驱动企业数据API编排技能,自动生成工具 + 编排多步工作流(工商四要素核验/工商查询/涉诉查询等)
Cron表达式解析、验证、可视化Schedule,以及常见定时任务模式生成
Git常见问题排查与修复,包括冲突处理、回退、清理、cherry-pick、reflog恢复等疑难杂症
根据描述或模板自动生成真实感测试JSON数据,支持faker模式、边界条件测试数据、负数/空值等异常数据构造
API contract testing: verify that API providers and consumers honor agreed-upon interfaces without running full integration. Use when building microservices, running Consumer-Driven Contract (CDC) tests, testing REST/GraphQL/Protobuf contracts, or ensuring frontend-backend API compatibility. Covers Pact, OpenAPI validation, GraphQL schema testing, and end-to-end contract suites.
Chaos engineering for distributed systems: design and run chaos experiments to proactively discover weaknesses before they cause outages. Use when user mentions chaos, fault injection, resilience testing, GameDays, chaos monkey,LitmusChaos, Gremlin, steady-state hypothesis, or when building信頼性 into microservices.
| name | research-agent |
| description | AI领域深度调研Agent——多源搜索+结构化输出+迭代深挖,适用于竞品分析、技术选型、市场调研。 |
| trigger | ["调研","research","竞品分析","技术选型","行业分析","市场调研","深度搜索"] |
对任意AI/技术主题执行结构化深度调研,输出可直接使用的研究报告。解决"需要快速了解某个领域但信息分散"的问题。
# [主题] 深度调研报告
## 📌 一句话结论
[核心发现]
## 🏢 竞品/项目概览
[列出主要玩家,一句话描述]
## 🔢 关键数据
- [数据点1]
- [数据点2]
- [数据点3]
## ✅ 核心优势
[3-5个核心优势]
## ⚠️ 局限性/风险
[2-3个主要问题]
## 📦 生态与社区
- Stars / 社区规模
- 文档质量
- 维护活跃度
## 💡 适合场景
[谁适合用,什么场景]
## 🔮 趋势判断
[未来3-6个月预判]
## 🔗 参考链接
[来源链接]
搜索角度:
- "框架名 + GitHub stars 2026"
- "框架名 vs 竞品 2026 comparison"
- "site:github.com 框架名"
- "框架名 official documentation"
搜索角度:
- "2026 AI tool release [month]"
- "AI product launch [category] 2026"
- "best [category] AI tools 2026"
搜索角度:
- "[技术A] vs [技术B] 2026"
- "when to use [framework]"
- "[framework] limitations problems 2026"
触发:用户提供调研主题
tool: llm-task(或内部分析)
分析主题,确定3-5个搜索角度:
tool: batch_web_search
{
"queries": [
{ "query": "[主题] GitHub 2026 stars features", "num_results": 8 },
{ "query": "[主题] vs [竞品] 2026 comparison", "num_results": 6 },
{ "query": "[主题] limitations problems 2026", "num_results": 6 },
{ "query": "[主题] use cases real world 2026", "num_results": 6 },
{ "query": "site:github.com [主题] 2026", "num_results": 5 }
]
}
tool: extract_content_from_websites
从搜索结果中选取2-3个高质量来源(如官方文档、权威评测)深度提取:
{
"tasks": [
{ "url": "<url1>", "prompt": "提取关键功能、优缺点、定价、使用限制", "task_name": "主站文档" },
{ "url": "<url2>", "prompt": "提取评测数据、用户反馈、对比信息", "task_name": "对比评测" }
]
}
tool: llm-task
将搜索结果合成结构化报告,填充输出框架的每个章节。
tool: write
{
"path": "memory/research/[主题]_[日期].md",
"content": "<完整报告内容>"
}
触发:用户对报告继续提问或要求深挖
在上一份报告基础上:
environment:
required_tools:
- batch_web_search # 多源搜索
- extract_content_from_websites # 深度内容提取
- llm-task # 搜索结果合成
- write # 报告存储
required_permissions:
- 网络访问
用户输入:
调研一下 2026年AI编程工具的格局,重点关注Claude Code竞品
Agent行为:
memory/research/ai_coding_tools_2026_0407.md