Token导航 LogoToken导航

Agent Skill · 研究检索

openclaw-skill-intelligence-ingestion(OpenClaw 技能 intelligence ingestion)

openclaw-skill-intelligence-ingestion 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

openclaw-skill-intelligence-ingestion 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为操作浏览器;许可证为 MIT-0,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 8,200。

总安装 16,944周安装 706GitHub Stars 公开资料未说明下载 5,648核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:openclaw-skill-intelligence-ingestion(OpenClaw 技能 intelligence ingestion)
来源仓库:https://github.com/2233admin/openclaw-skill-intelligence-ingestion
安装命令:
openclaw skills install openclaw-skill-intelligence-ingestion
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install openclaw-skill-intelligence-ingestion

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

自动分析 URL 内容并生成黑曜石笔记更新内存。适合快速捕获外部信息与战略价值点。支持文章、推文等多种内容类型接入。需配置正确的 API 密钥与访问权限。注意隐私数据过滤与存储合规要求。openclaw-skill-intelligence-ingestion 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install openclaw-skill-intelligence-ingestion;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

openclaw-skill-intelligence-ingestion 怎么安装?

在终端执行 openclaw skills install openclaw-skill-intelligence-ingestion。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

openclaw-skill-intelligence-ingestion 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

openclaw-skill-intelligence-ingestion 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

openclaw-skill-intelligence-ingestion 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

openclaw-skill-intelligence-ingestion 用的人多吗?

公开数据是累计安装 16,944 次,近一周安装 706 次,下载 5,648 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

openclaw-skill-intelligence-ingestion 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
openclaw-skill-intelligence-ingestion

OpenClaw 技能 intelligence ingestion · 本页

16,944—操作浏览器MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
Intelligence Ingestion
description
Auto-analyze URLs/info for OpenClaw strategic value, classify, create Obsidian notes, update memory. Use when user shares a URL, article, tweet, or any external info source.

Intelligence Ingestion Skill

When the user shares a URL, article, tweet, thread, or any piece of external information, execute this pipeline automatically. Do NOT ask for permission — just process it.

Trigger Conditions

USE this skill when:

  • User sends a URL (x.com, github.com, any domain)
  • User pastes article text or tweet content
  • User says "analyze this", "evaluate this", "what do you think about this"
  • User forwards content from Telegram or any messaging surface
  • User shares something and asks about strategic value

DO NOT USE when:

  • User is asking a general question unrelated to external content
  • User is asking about internal workspace files
  • User explicitly says they don't want analysis

Pipeline Steps

Step 1: READ — Extract Content

  1. If a URL is provided, read the full content (use read_url_content or browser)
  2. If text is pasted, use it directly
  3. If X/Twitter link fails to parse (common), search web for the tweet content

Step 2: CLASSIFY — What Is This?

Assign one primary category and up to 2 secondary tags:

CategoryDescriptionExamples
infraInfrastructure / protocols / networkingPilot Protocol, MCP, networking stacks
strategyRouting / cost / architecture decisionsModel routing, multi-account, fallback chains
skillAgent skills / tools / capabilitiesOpenAI Skills, Skill design patterns
theoryConceptual frameworks / mental modelsBayes, decision theory, learning loops
tutorialHow-to guides / learning materialClaude Code, OpenClaw tutorials
productTools / apps / servicesLM Studio, new AI models, apps
communityOpenClaw ecosystem / discussionsCommunity posts, feature requests
threatRisks / security / deprecationAPI changes, breaking updates, security alerts

Step 3: ANALYZE — Strategic Value Assessment

For each piece of information, evaluate:

## Strategic Assessment
- **What is it?** [One sentence]
- **What can it do for us?** [Specific capability/benefit]
- **What can we build with it?** [Concrete output/project]
- **Strategic value:** [🔴 Critical / 🟡 High / 🟢 Medium / ⚪ Low]
- **Competitive edge:** [What advantage over people who don't have this?]
- **Relation to active bottleneck:** [Does it relate to context overflow/token saving?]

Reference the current engineering bottleneck from MEMORY.md → "Active Engineering Bottleneck" section.

Step 4: MAP — Relate to Existing Architecture

Check against the OpenClaw stack:

SOUL.md → PRINCIPLES.md → AGENTS.md (Identity Stack)
MEMORY.md (System State + Bottleneck)
TOOLS.md (Coprocessors: Codex, Antigravity, LM Studio)
Pilot Protocol (Context Separation Layer - P0)

Determine which layer this information impacts and note dependencies/synergies with existing components.

Step 5: STORE — Write Obsidian Note

Create a note at:

/Volumes/T7 Shield/Obsidian_Vault/20_Intelligence/YYYYMMDD_AuthorOrSource_ShortTitle.md

Follow this template exactly:

# [Title]

**Source:** [Link](URL)
**Date:** YYYY-MM-DD
**Category:** [Primary Category] / [Secondary Tags]
**Strategic Value:** [🔴/🟡/🟢/⚪] + one-line reason
**Relation to Active Bottleneck:** [Yes/No + how]

---

## Summary
[2-3 paragraphs of core content]

## Key Takeaways
[Numbered list of actionable insights]

## Impact on OpenClaw Architecture
[How this relates to our stack]

## Action Items
[What to do next, if anything]

---

**Keke's Note:** [Opinionated analysis in Keke's voice — direct, no-BS, relate to 阳哥's goals]

Step 6: REMEMBER — Update Memory

  1. Always append to today's daily log: ~/.openclaw/workspace/memory/YYYY-MM-DD.md
  2. If strategic value is 🔴 Critical: Also update MEMORY.md (Pending Work or Active Bottleneck)
  3. If it suggests a new principle: Flag for potential PRINCIPLES.md update
  4. If it's a new tool/service: Flag for potential TOOLS.md update

Step 7: RESPOND — Summarize to User

Reply with a concise summary:

📥 已摄取: [Title]
📂 类别: [Category]
🎯 战略价值: [🔴/🟡/🟢/⚪] [One-line reason]
💾 已存档: Obsidian → 20_Intelligence/[filename]
📝 已记录: memory/YYYY-MM-DD.md
🔗 关联: [Which existing component it relates to]
⚡ 建议行动: [Next step, if any]

Edge Cases

  • Multiple URLs in one message: Process each separately, create separate Obsidian notes
  • Duplicate/similar content: Check if similar note exists, merge or reference instead of duplicating
  • Non-English content: Analyze in original language, write notes in Chinese (matching existing vault style)
  • Paywalled/inaccessible content: Note as "content unavailable" and work with whatever user provided
  • User provides their own analysis: Incorporate their judgment, don't overwrite — they know their system best

Quality Checklist

Before completing, verify:

  • [ ] Obsidian note created with correct filename format
  • [ ] Daily memory log updated
  • [ ] Source URL preserved in note
  • [ ] Strategic value assessed against current bottleneck
  • [ ] Keke's Note written with genuine analysis (not generic)
  • [ ] User received confirmation summary

平台分布

OpenClaw

93.71%
按下载量换算5,293

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

未展示

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills