agentCLAWHUBUnverified

Industry Deep Dive Pipeline

This skill should be used when turning a topic brief, research materials, vendor case, policy event, or industry question into a publish-ready single deep-di... Skill: Industry Deep Dive Pipeline Owner: haiyangchenbj Summary: This skill should be used when turning a topic brief, research materials, vendor case, policy event, or industry question into a publish-ready single deep-di... Tags: latest:1.1.0 Version history: v1.1.0 | 2026-10-08T04:07:14.217Z | user Align to v1.1.0: add not_for scope guard, sharpen description/description_zh v1.0.5 | 2026-09-14T02:35:47.655Z | user

OpenClaw

Rank

62

Safety

84

Downloads

956

Updated

Oct 12, 2026

Version

1.1.0

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 956 downloads reported by the source. Last updated 10/12/2026.

Avoid when

  • Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor
Clawhubvendor · observed Oct 12, 2026
Protocol compatibility
OpenClawcompatibility · observed Oct 12, 2026
Adoption signal
956 downloadsadoption · observed Oct 12, 2026
Latest release
1.1.0release · observed Oct 8, 2026
Handshake status
UNKNOWNsecurity

What changed

Newest first. Only changes that point at a source are listed.

  • Release 1.1.0release

    Align to v1.1.0: add not_for scope guard, sharpen description/description_zh

    Observed Oct 8, 2026 · source

Install and run

Setup complexity: low.

clawhub skill install s17672gh0nx9qr7sjp84kz7xen83jv7v:industry-deep-dive-pipeline-skill
  1. Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
  2. Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-haiyangchenbj-industry-deep-dive-pipeline-skill/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

147,772 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
slug: industry-deep-dive-pipeline-skill
displayName: Industry Deep-Dive Pipeline
not_for:
  - Quick single-pass summaries or news recaps (multi-stage research pipeline)
  - Marketing copy, product one-pagers, or social posts (long-form article output)
  - Publishing or distributing the finished article (stops at publish-ready draft)
  - Research without a topic brief or source materials (needs an input corpus)
name: industry-deep-dive-pipeline
description: >
  This skill should be used when turning a topic brief, research materials,
  vendor case, policy event, or industry question into a publish-ready single
  deep-dive article for technology, AI, data, cloud, or enterprise-software
  audiences. It runs source verification, originality and competition review,
  full editorial planning, two human decision gates, drafting, deterministic
  red-line checks, an existing eight-role review panel, revision, and final
  evidence packaging. It stops at an approved Markdown article plus evidence and
  review records; it does not create publication layouts, covers, social copy,
  CMS drafts, or publish content.

  【适用】中立第三方产业深度研究 / 行业长文(个人 IP、公众号深度稿、厂商案例的独立分析)。【不适用】品牌营销稿、产品稿、按 content brief
  写的推广文、技术教程——这类内容请用对应项目工作区里专门的写作 Skill;本 Skill 不做品牌露出与营销措辞,也不产出发布物料。
description_zh: 产业深度文章全流程编排
description_en: Industry deep-dive pipeline
version: "1.1.0""
agent_created: true
read_when:
  - 产业深度文章
  - 行业深度稿
  - 从选题到定稿
  - 原创性复核与写作
  - 科技行业长文
  - industry deep dive
  - research article pipeline
---

# Industry Deep-Dive Pipeline

将单篇科技、AI、数据、云和企业软件产业选题推进到审核通过的正式 Markdown。采用 Prompt Chaining + Evaluator-Optimizer:固定主链保证可追踪,八角色会审后的修订循环保证质量。

## When to use

- 输入包含选题简报、事件、厂商案例、政策变化或研究材料,需要形成有判断、有证据、有边界的深度文章。
- 需要先查中文与英文同类内容,确认信息增量和表述撞车风险。
- 需要在正文前完成目标读者、核心判断、论证链、反方、结构、标题和风险策划。
- 需要事实表、原创性复核、会审记录和终稿复检形成完整证据包。

## Do not use

- 新闻快讯、摘要、产品说明书、技术教程或营销稿。
- 厂商大会全景报告;改用 `vendor-summit-report`。
- LinkedIn 单平台短观察;改用 `linkedin-industry-observation`。
- 微信排版、封面、摘要、关键词、社交物料、CMS/Notion 入库或发布;交给独立发布套件。
- 多篇系列的自动排期与批量生成。系列信息仅作单篇上下文。

## Required inputs

读取 `references/planning-schema.md`,建立 case bundle。至少确认:

- `topic`
- `source_materials`
- `target_reader`
- `target_format`
- `length_range`
- `writing_profile`

可选输入:合集或系列关系、历史平台信号、必须处理的反方、本篇新增红线、时效窗口和用户已有假设。

## Workflow

### Step 1: [Deterministic] Diagnose input

1. 校验路径、URL、日期、目标格式和 writing profile。
2. 判断文章类型、时效等级、事实风险与敏感性。
3. 把用户已有判断标为待验证假设,不直接当成结论。
4. 复制 `templates/case-brief.template.md` 为 `00-case-brief.md`。
5. 运行:

```bash
python scripts/validate_case_bundle.py --case-dir <case-dir> --stage input --enforce
```

输入门禁失败时停止,报告缺项。

### Step 2: [LLM] Build fact table

1. 从材料提取数字、日期、公司、政策、产品状态、引述和因果声明。
2. 使用搜索发现来源,回到官网、财报、论文、监管原文或官方发布核验关键事实。
3. 高风险事实至少双源确认,或由一手权威源确认。
4. 记录原始日期、核验日期、状态、来源等级、可使用措辞和失效条件。
5. 复制 `templates/fact-table.template.md` 为 `01-fact-table.md`。
6. 按需读取 `references/evidence-and-originality.md`。

### Step 3: [LLM] Review 

_meta.json

{
  "ownerId": "kn70yg6zwmkftx4939qrs89awx82rr9a",
  "slug": "industry-deep-dive-pipeline-skill",
  "version": "1.1.0",
  "publishedAt": 1791432434217
}

references/evidence-and-originality.md

# Evidence and Originality

## Source tiers

| 等级 | 来源 | 用法 |
|---|---|---|
| A | 官方公告、监管原文、财报、论文、产品文档、法院文件、公开演讲原文 | 可支撑高风险事实 |
| B | Reuters、Bloomberg、FT 等高可信二手媒体,或有明确采访对象的专业媒体 | 可交叉确认;缺一手时标注来源边界 |
| C | 行业分析、播客、博客、社区文章 | 用于观点、竞争覆盖和反方;数字需回溯 |
| D | 社交媒体、转载、搜索摘要、截图 | 仅作线索或待核信号 |

## Fact table status

- `confirmed`:一手源确认,或两个独立可靠来源一致。
- `qualified`:事实成立,但需保留来源、时间或 vendor-reported 限定。
- `contested`:存在直接冲突,正文必须呈现边界或不使用。
- `pending`:尚未完成核验,禁止进入终稿。
- `rejected`:已确认错误、过时或与文章无关。

## Fact table minimum fields

| ID | Claim | Type | Status | Original date | Verified at | Source tier | Source URL/path | Allowed wording | Expiry/invalidating signal |
|---|---|---|---|---|---|---|---|---|---|

类型包括:number、date、status、quote、policy、market、causal、interpretation。

## High-risk rules

1. 政策和法律状态明确区分讨论、征求意见、草案、通过、生效和执行案例。
2. 产品状态明确区分 roadmap、private preview、public preview、GA 和 deprecated。
3. 融资区分本轮融资额、累计融资额、估值和交易对价。
4. 厂商性能数字标注 vendor-reported;客户效果标注 customer-reported。
5. 宏观数据保留统计主体、时间、分母和单位。
6. 高时效主题在写作当天或终稿前重新核验。

## Originality review

### Search units

分别搜索:

- 主题与主体。
- 核心判断的同义表达。
- 分析框架与关键区分。
- 标志性标题和记忆句。
- 直接反方和历史反例。
- 中文、英文及主题相关语言。

### Competition categories

| 类型 | 判断 |
|---|---|
| news coverage | 事实已报道,不代表分析撞车 |
| profile/financing story | 公司与人物叙事充分,不能再以介绍为主线 |
| same topic, different thesis | 可写,但需明确增量 |
| same thesis, different evidence | 空间有限,需更强框架或新证据 |
| same thesis and structure | 高撞车风险,换角度或放弃 |

### Allowed conclusions

- “现有覆盖主要集中在 X,本文增量位于 Y。”
- “该判断已有公开讨论,本文补充了 Z 证据或边界。”
- “未检索到将 A、B、C 组合为同一框架的中文深度分析;结论受当前检索范围限制。”

### Forbidden shortcuts

- 只搜一个关键词就断言“中文无人写”。
- 把标题不同当作观点不同。
- 把用户首次提出当作首创。
- 用“有媒体报道”替代文章和论证的逐项比较。
- 忽略直接反方,再用单向证据证明结论。

## Stop conditions

- 核心判断和论证结构均已被充分覆盖。
- 差异化依赖无法核实或明显错误的事实。
- 唯一增量是更夸张的标题。
- 主题高风险且缺少可靠来源。

停止后输出:放弃、并入其他文章、缩为短观察、等待新证据或更换角度。

references/planning-schema.md

# Planning Schema

## Case bundle

每篇文章使用独立 case 目录。目录名建议为 `YYYY-MM-DD-topic-slug`,不包含敏感客户名、账号或数据库 ID。

## 00-case-brief.md

必填字段:

| 字段 | 允许值或说明 |
|---|---|
| topic | 一个可回答的问题或明确主题 |
| source_materials | 本地路径与 URL 列表 |
| target_reader | 核心层、扩展层、推荐层;至少核心层明确 |
| target_format | 正式 Markdown 长文 |
| length_range | 最小与最大字数 |
| writing_profile | JSON/YAML 路径或 `generic` |
| timeliness | low / medium / high |
| confidentiality | public / private-personal / private-company |

可选字段:collection_context、platform_signals、must_address、red_lines、deadline、user_hypothesis。

## Risk classification

| 风险 | 触发条件 | 处理 |
|---|---|---|
| high | 政策、法律状态、融资估值、产品 GA/Preview、宏观数字、原创/唯一声明 | 一手源或双源;记录核验时间 |
| medium | 厂商定位、市场份额、客户案例、性能数据 | 原始出处优先;标注 vendor/customer-reported |
| low | 通用背景、公开定义、解释性类比 | 至少一个可靠来源或由文章明确标为解释 |

## Planning brief required sections

1. 选题可行性与竞争程度。
2. 事实校准与必须修正的原始判断。
3. 目标读者分层及阅读任务。
4. 核心判断、论证链与边界。
5. 原创性锚点和现有覆盖缺口。
6. 反方、历史反例和证伪变量。
7. 标题方向、摘要功能和标题风险。
8. 结构大纲与字数分配。
9. 收藏载体:表格、框架、时间线或对比。
10. 本篇红线、时效要求和敏感性。
11. 与历史文章或系列的关系。
12. Gate A 待确认项。

## Gate A record

```yaml
gate_a:
  status: approved | changes_requested | pending
  approved_at: ISO-8601
  decisions:
    thesis: ""
    title_direction: ""
    length_and_depth: ""
    controversy_handling: ""
    platform_and_series: ""
  rejected_options: []
```

## Gate B record

```yaml
gate_b:
  status: approved | not_required | pending
  approved_at: ISO-8601
  tensions:
    - issue: ""
      options: []
      decision: ""
      reason: ""
```

## Completion rules

- Gate A 为 pending 时最多完成 `03-planning-brief.md`。
- Gate B 为 pending 时最多完成 `06-review-record.md`。
- `07-final.md` 生成后必须存在终稿机器扫描和 `08-final-check.md`。
- 发布物料字段不得进入 case bundle;后续发布 Skill 另行读取 `07-final.md`。

references/replay-evaluation.md

# Replay Evaluation

## General method

历史回放用于验证流程能否复现关键决策与门禁,不重新评价作者当时的判断。回放只读历史文件,不修改原稿、策划和发布物料。

每个案例记录:

- 历史输入和最终结果。
- Skill 预期生成的 case bundle。
- Gate A 应出现的确认项。
- Gate B 应出现的张力项。
- 事实与红线应被哪一道门禁拦截。
- 与历史流程的差异和遗漏。

## Case 1: Fireworks AI

### Inputs

- `Fireworks-AI-策划-20260727-v1.md`
- `Fireworks-AI-策划-20260727.md`
- `Fireworks-AI-正文初稿-20260727.md`

### Must reproduce

- Fireworks 作为案例证据,文章主线是推理基础设施层重估。
- 原创性复核识别已有融资、人物和 Token 工厂报道。
- 正面处理纯托管无结构性护城河的反方。
- Gate A 覆盖标题、compute-to-data 比重、中美对比深���、反方引用和 Cursor 位置。
- 终稿保留 Cursor 作为证伪变量。

### Failure signals

- 写成公司介绍或融资复盘。
- 把“推理即护城河”当作无需证明的前提。
- 忽略 SemiAnalysis 张力或客户集中度。
- 把摘要、标签和发布备注混入正式正文。

## Case 2: Kimi / model-weight controls

### Inputs

- `中国AI出口管制-策划-20260727.md`
- `中国AI出口管制-正文初稿-20260727.md`

### Must reproduce

- 把“已启动”修正为讨论、酝酿或未落地。
- 修正宏观数据分母和时间口径。
- 检索 2025 年美国模型权重管制及撤销的历史。
- 形成权重与 Token、资产与服务、能力定价与合规定价的分析框架。
- Gate A 覆盖标题、蒸馏指控篇幅、合集归属和发布时间。
- 终稿前重新核验政策状态。

### Failure signals

- 把搜索摘要当作政策原文。
- 混用讨论、草案和正式禁令。
- 滑向地缘输赢叙事。
- 未处理法律状态和时效变化。

## Case 3: Europe AI — Open Source as Sovereignty

### Inputs

- `文章_微信_01_开源即主权.md`
- `欧洲AI深度研究系列_三平台传播运营策划.md`
- `审核报告_欧洲AI深度研究系列_全面复审.md`

### Must reproduce

- 系列定位只作为上下文,单篇可以独立阅读。
- 三平台共享事实与核心判断,平台表达分别适配。
- 识别中英文对举、系列自指、元信息、营销词和口语缓冲。
- 改稿后执行机器扫描与人工通读。
- 不把系列排期、运营动作和商业意图写进正文。

### Failure signals

- 把“前几篇、下一篇、系列收官”等过程信息留在单篇。
- 中文规则严格、英文规则放松。
- 只跑一次 Grep,改稿后不复扫。
- 把运营 KPI 当作文章论据。

## Acceptance record

```markdown
# Replay Result

## Case
- Name:
- Historical inputs:
- Read-only confirmed: yes/no

## Workflow coverage
- Input gate:
- Fact table:
- Originality review:
- Gate A:
- Draft gate:
- Eight-role review:
- Gate B:
- Final re-check:

## Differences
- Expected but missed:
- Newly added but unnecessary:
- False positives:
- False negatives:

## Verdict
- pass / conditional pass / fail
- Required skill changes:
```

第一版验收要求三个案例均达到 pass 或 conditional pass,且无 P0 安全问题、无历史文件修改。
Github ReposUpdated 3d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW
Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW
Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.

MCPOPENCLAW
Github ReposUpdated 7mo agoRank 70

CopilotKit

The Frontend for Agents & Generative UI. React + Angular

OPENCLAW

Machine-readable data

The same record, as JSON, for agents and crawlers.

{
  "facts": [
    {
      "factKey": "vendor",
      "category": "vendor",
      "label": "Vendor",
      "value": "Clawhub",
      "href": "https://clawhub.ai/haiyangchenbj/skills/industry-deep-dive-pipeline-skill",
      "sourceUrl": "https://clawhub.ai/haiyangchenbj/skills/industry-deep-dive-pipeline-skill",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-12T02:13:10.933Z",
      "isPublic": true
    },
    {
      "factKey": "protocols",
      "category": "compatibility",
      "label": "Protocol compatibility",
      "value": "OpenClaw",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-haiyangchenbj-industry-deep-dive-pipeline-skill/contract",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-haiyangchenbj-industry-deep-dive-pipeline-skill/contract",
      "sourceType": "contract",
      "confidence": "medium",
      "observedAt": "2026-10-12T02:13:10.933Z",
      "isPublic": true
    },
    {
      "factKey": "traction",
      "category": "adoption",
      "label": "Adoption signal",
      "value": "956 downloads",
      "href": "https://clawhub.ai/haiyangchenbj/industry-deep-dive-pipeline-skill",
      "sourceUrl": "https://clawhub.ai/haiyangchenbj/industry-deep-dive-pipeline-skill",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-12T02:13:10.933Z",
      "isPublic": true
    },
    {
      "factKey": "latest_release",
      "category": "release",
      "label": "Latest release",
      "value": "1.1.0",
      "href": "https://clawhub.ai/haiyangchenbj/industry-deep-dive-pipeline-skill",
      "sourceUrl": "https://clawhub.ai/haiyangchenbj/industry-deep-dive-pipeline-skill",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-10-08T04:07:14.217Z",
      "isPublic": true
    },
    {
      "factKey": "handshake_status",
      "category": "security",
      "label": "Handshake status",
      "value": "UNKNOWN",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-haiyangchenbj-industry-deep-dive-pipeline-skill/trust",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-haiyangchenbj-industry-deep-dive-pipeline-skill/trust",
      "sourceType": "trust",
      "confidence": "medium",
      "observedAt": null,
      "isPublic": true
    }
  ],
  "events": [
    {
      "eventType": "release",
      "title": "Release 1.1.0",
      "description": "Align to v1.1.0: add not_for scope guard, sharpen description/description_zh",
      "href": "https://clawhub.ai/haiyangchenbj/industry-deep-dive-pipeline-skill",
      "sourceUrl": "https://clawhub.ai/haiyangchenbj/industry-deep-dive-pipeline-skill",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-10-08T04:07:14.217Z",
      "isPublic": true
    }
  ]
}

Record generated Oct 12, 2026.

For crawlers

This page is free to read. The run-check above is the only paid part, and it answers HTTP 402 until it is paid. Everything else here is public.

  • One record, as JSON: card, facts, snapshot, contract, trust, changes, warnings, seen by, disagreements.
  • Every agent, one feed: /.well-known/ai-catalog.json
  • What this site sells, and the price: /.well-known/x402
  • Paid run-check: /api/v1/agents/clawhub-haiyangchenbj-industry-deep-dive-pipeline-skill/run-check

Sponsored

Ads related to Industry Deep Dive Pipeline and adjacent AI workflows.