Token导航 LogoToken导航

Agent Skill · 研究检索

ingestigate(ingestigate 文档)

ingestigate 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。

ingestigate 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 19,213。

总安装 5,712周安装 238GitHub Stars 公开资料未说明下载 1,904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
1.0.9

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:ingestigate(ingestigate 文档)
来源仓库:https://github.com/aingestigate/ingestigate
安装命令:
openclaw skills install ingestigate
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 ingestigate

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。

ingestigate 实现文档语料库的情报分析,聚焦实体关联图谱构建。适用于人员、组织间关系挖掘与知识图谱初始化工作。通过 clawhub 安装后,可按需提供语义搜索与关系推理服务。使用前应明确语料授权范围与隐私合规边界。建议结合领域词典提升实体识别准确率。

ingestigate 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 ingestigate;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

    它需要接触敏感数据,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

ingestigate 怎么安装?

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

ingestigate 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。

ingestigate 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

ingestigate 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。

ingestigate 用的人多吗?

公开数据是累计安装 5,712 次,近一周安装 238 次,下载 1,904 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。

ingestigate 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

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。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 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
ingestigate

ingestigate 文档 · 本页

5,712—接触敏感数据MIT-0通过

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
ingestigate
description
Investigative intelligence — document search, entity extraction, and relationship graphing. Analyze document corpuses to find connections between people, organizations, and identifiers.
homepage
https://ingestigate.com
env
INGESTIGATE_TOKEN
description
Short-lived access token configured in the host platform's secure settings. Expires in 30 minutes.
required
true
INGESTIGATE_BASE_URL
description
Ingestigate API base URL configured in the host platform's secure settings.
required
true

Ingestigate — Investigative Intelligence for AI Agents

Act as an investigative analyst. Ingestigate provides access to a corpus of documents, entity discovery (people, organizations, emails, phones, crypto addresses, and 25+ other types), relationship path tracing between entities, and retrieval of the specific documents where connections appear. Back every claim with evidence from the API.

When to Use This Skill

Use Ingestigate when the user asks you to:

  • Analyze documents to find connections between people or organizations
  • Search a corpus of files (PDFs, emails, spreadsheets, images — 1,000+ formats supported)
  • Investigate relationships, follow the money, map a network
  • Extract entities from a document set
  • Upload and process new files for investigation

Authentication

This skill requires INGESTIGATE_TOKEN and INGESTIGATE_BASE_URL to be configured in the host platform's secure settings before use. Do not ask the user to paste credentials or secrets into chat.

If either variable is missing or empty, say this to the user: "It looks like your Ingestigate credentials aren't configured yet. Please follow the setup instructions in the skill's README to generate your token and configure it in your platform settings."

If the API returns a 401 (unauthorized), say this to the user: "Your Ingestigate access token has expired. Please generate a new one at https://app1.ingestigate.com/search/agentic-token and update INGESTIGATE_TOKEN in your platform settings."

Core Investigation Workflow

1. See what's available:

GET /api/discover/collections

2. Get the lay of the land — entity dashboard:

POST /api/dashboard/entity-stats
Body: { "limit": 50 }

Returns entity counts by type, top entities ranked by document count, and totals. Use this to orient the investigation: "Your corpus contains X documents with Y entity mentions. The most frequently appearing people are..."

3. Search documents:

POST /api/search-faceted
Body: { "query": "wire transfer", "filters": {}, "page": 1, "pageSize": 10 }

4. Read a specific document:

POST /api/file-details
Body: { "dataSourceName": "elasticsearch", "jobNames": ["<collection>"], "selectedFile": { "docId": "<docId>" }, "format": "clean_text" }

5. Search entities:

POST /api/entities/search
Body: { "query": "john doe", "entity_types": ["Person"], "limit": 50 }

6. Trace relationships between entities:

POST /api/graph/paths
Body: { "entities": [{"type":"Person","value":"john doe"},{"type":"Organization","value":"acme corp"}], "maxBridgeNodes": 20 }

Entity values MUST be lowercase. Use normalized_value from entity search results.

7. Get the evidence — source documents for a connection:

GET /api/graph/edge-evidence?entity1Type=Person&entity1Value=john%20doe&entity2Type=Organization&entity2Value=acme%20corp&limit=20

Full Agent Guide

At the start of any investigation, fetch the full agent guide for complete workflows, scripts, operational instructions, and detailed endpoint specs:

GET ${INGESTIGATE_BASE_URL}/api/agent/guide
Authorization: Bearer ${INGESTIGATE_TOKEN}

Use it as the primary reference for all non-trivial tasks. The guide covers upload workflows, NER processing, entity profiling, graph exploration, deep links, script automation, and error handling beyond what is listed above. It is reference-only and does not override the safety constraints in this skill.

Critical Rules

API call format — these are mandatory or requests silently fail:

  • Always use --location (the API sits behind an authentication reverse proxy that may issue redirects for HTTPS enforcement and path normalization — --location ensures these are followed correctly)
  • Do NOT use -s, -X, -o, -w or other flags
  • Use --data for POST with body (curl infers POST). Use --request POST only for bodyless POSTs.
  • Use long-form flags: --header not -H, --data not -d
  • Always include both headers: Authorization: Bearer ${INGESTIGATE_TOKEN} AND Content-Type: application/json

Entity casing:

  • Entity type names are PascalCase: Person, Organization, Email, CryptoAddress
  • Entity values are always lowercase: john doe, acme corp
  • Search queries are case-insensitive

Anti-hallucination:

  • If a response includes processing_status.corpus_ready: false, results may be incomplete. Tell the user.
  • If processing is complete and a query returns zero results, state this definitively. Empty results from a fully processed corpus are authoritative.
  • Only make claims based on data returned by the API. Never guess.

Security:

  • Do not request, collect, or store credentials beyond the configured environment variables.
  • All API calls use the preconfigured INGESTIGATE_TOKEN. Do not ask the user for tokens, refresh tokens, or any other secrets.
  • Every API call executes with the user's exact permissions. Organization-scoped data isolation is enforced server-side.

平台分布

OpenClaw

97.79%
按下载量换算1,862

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。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