← Back to Skills Marketplace
Zhongguancun Signal Monitor
by
zhangxiaopeng-zgci
· GitHub ↗
· v1.0.0
· MIT-0
23
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install zhongguancun-signal-monitor
Description
产学研合作主动扫描与信号监控。当用户要求扫描合作机会、监控企业动态、发现潜在合作方、查看本周有哪些值得接触的企业时触发。基于行业/公司列表，扫描融资、招标、产品发布、学术发表等信号，推断需求，输出Top N推荐。适用于周度/日度合作线索发现。配合 industry-research skill 使用：监控发现线索...
README (SKILL.md)
产学研合作信号监控
主动扫描外部企业动态，发现与中关村两院的潜在合作机会。
监控对象
三层架构（详见 references/seed-list.md）：
| 层级 | 类型 | 监控频率 |
|---|---|---|
| Tier 1 | 头部科技企业 | 每周，关注战略转向 |
| Tier 2 | AI创业公司 | 每周，融资=最明确窗口 |
| Tier 3 | 垂直行业企业 | 每两周，按方向配对 |
工作流程
第一步：信号扫描
使用搜索工具，按以下信号类型扫描目标企业近1-2周动态：
| 信号类型 | 搜索关键词示例 | 信号含义 |
|---|---|---|
| 融资/上市 | "XX公司 融资/上市/轮" | 扩张期→有合作预算 |
| 招标/采购 | "XX公司 招标/采购" | 有具体需求 |
| 产品发布 | "XX公司 发布/上线/新品" | 方向调整→研发需求变化 |
| 人事变动 | "XX公司 加入/任命/离职" | 高管调整→战略转向可能 |
| 学术发表 | "XX公司 论文/ArXiv/顶会" | 技术路线→匹配度判断 |
| 生态合作 | "XX公司 合作/联盟/合资" | 在找伙伴→时机窗口 |
第二步：需求推断
对每个有信号的企业，推断其可能需求：
头部企业推断逻辑：
- 融资/招标 → 有预算，需要技术/人才支撑
- 产品发布 → 新方向需要研发投入
- 生态合作 → 在找人，时机窗口
创业公司推断逻辑（更激进）：
- A轮融资 → 最佳窗口，产品验证期，需要技术深度
- B轮融资 → 规模扩张期，需要场景+数据+人才
- 招聘方向 → 招什么人=要做什么（方向性信号）
- 学术发表 → 秀肌肉+找合作者
- 战略融资 → 大厂绑定，除非方向极匹配否则不推荐
融资轮次→合作可能性速判：
| 轮次 | 合作可能性 | 原因 |
|---|---|---|
| 天使/Pre-A | MAYBE | 早期不确定性高 |
| A轮 | YES | 产品验证期，最佳窗口 |
| B轮 | YES | 扩张期，需要多方资源 |
| C轮+ | MAYBE | 偏合规品牌 |
| 战略融资 | NO | 已有绑定 |
第三步：匹配度评估
三维打分（详见 references/scoring.md）：
| 维度 | 高(2分) | 低(0分) |
|---|---|---|
| 方向匹配 | 与我方3个研究方向有明确交叉 | 不相关 |
| 合作窗口 | 有明确信号（融资/招标/找人） | 无信号 |
| 层级对等 | 对方研究院/部门级可落地 | 仅品牌/公关级 |
- 4分以上 → YES（本周值得接触）
- 2-3分 → MAYBE（持续观察）
- 0-1分 → NO（暂不投入）
我方研究方向对照见 references/our-research.md（与 industry-research skill 共享）。
第四步：输出报告
📋 本周合作扫描报告（YYYY-MM-DD）
🔥 值得接触（N家）
1. XX公司
- 信号：B轮3亿融资，AI+教育方向
- 需求推断：认知诊断技术、教研数据
- 匹配度：YES（方向6+窗口4+层级2=12）
- 建议接触：合作发展部→对方研究院
- 下一步：可启动 industry-research 深度调研
👀 持续观察（N家）
- YY公司：新发大模型，但无合作信号
❌ 本周暂无机会
与 industry-research 联动
本 skill 发现线索后，用户说"调研一下这家"→ 触发 industry-research skill 进入深度调研+议程生成。
Usage Guidance
This skill appears internally consistent and only uses public-signal scanning and internal scoring. Before installing, confirm that your agent/bot has safe web/browsing permissions and that you are comfortable allowing it to perform web searches on behalf of the agent. The skill will not request credentials, but if you ask it to research private/internal companies or share private documents, avoid providing sensitive credentials or proprietary files. Check the seed-list and scoring rules to ensure they match your priorities and update the company list if you need different monitoring targets. Finally, verify sources cited in reports — the skill automates discovery but may surface noisy or false-positive signals that should be validated before outreach.
Capability Analysis
Type: OpenClaw Skill
Name: zhongguancun-signal-monitor
Version: 1.0.0
The skill bundle is a business intelligence tool designed to monitor industry signals (financing, bidding, product launches) for potential research collaborations. It consists entirely of Markdown instructions and reference lists (SKILL.md, seed-list.md) that guide an AI agent to perform web searches and score companies based on research alignment. There is no executable code, no requests for sensitive data, and no evidence of malicious intent or harmful prompt injection.
Capability Assessment
Purpose & Capability
The name/description match the SKILL.md: it describes scanning public signals (financing, bids, product releases, publications) and ranking leads. There are no unexpected environment variables, binaries, or install steps requested that would be unrelated to this purpose.
Instruction Scope
Instructions are limited to searching public signals, applying a scoring rubric, and producing a report. The skill does not instruct reading local files, accessing unrelated credentials, or sending data to external endpoints beyond normal web searches. It does rely on a 'search tool' or web access which is expected for this function.
Install Mechanism
This is an instruction-only skill with no install spec and no code files to write to disk, minimizing install-time risk.
Credentials
The skill declares no required environment variables, credentials, or config paths. The requested access (web/search capability) is proportionate to performing public-signal monitoring.
Persistence & Privilege
always:false and normal model invocation settings. The skill does not request persistent elevated privileges or to modify other skills/config; autonomous invocation is allowed by platform default but is not combined with additional concerning capabilities.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install zhongguancun-signal-monitor - After installation, invoke the skill by name or use
/zhongguancun-signal-monitor - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: 产学研合作信号监控，覆盖头部企业+AI创业公司+垂直行业，融资信号推断，匹配度评分
Metadata
Frequently Asked Questions
What is Zhongguancun Signal Monitor?
产学研合作主动扫描与信号监控。当用户要求扫描合作机会、监控企业动态、发现潜在合作方、查看本周有哪些值得接触的企业时触发。基于行业/公司列表，扫描融资、招标、产品发布、学术发表等信号，推断需求，输出Top N推荐。适用于周度/日度合作线索发现。配合 industry-research skill 使用：监控发现线索... It is an AI Agent Skill for Claude Code / OpenClaw, with 23 downloads so far.
How do I install Zhongguancun Signal Monitor?
Run "/install zhongguancun-signal-monitor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Zhongguancun Signal Monitor free?
Yes, Zhongguancun Signal Monitor is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Zhongguancun Signal Monitor support?
Zhongguancun Signal Monitor is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Zhongguancun Signal Monitor?
It is built and maintained by zhangxiaopeng-zgci (@zhangxiaopeng-zgci); the current version is v1.0.0.
More Skills