Token导航 LogoToken导航

Agent Skill · 研究检索

insforge(insforge 数据库)

insforge 用于记录任务执行中的错误、用户纠正、经验和能力缺口,适合在 Codex、Claude、Cursor、Gemini CLI 中希望让 Agent 持续沉淀问题、修正和最佳实践时使用。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。

insforge 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为接触敏感数据;许可证信息还不明确,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 836。

总安装 155,976周安装 6,740GitHub Stars 13下载 54,672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
unknown

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:insforge(insforge 数据库)
来源仓库:https://github.com/insforge/agent-skills
仓库路径:skills/insforge
安装命令:
npx skills add https://github.com/insforge/agent-skills --skill insforge
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 npx skills 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。

skills.shnpx skills
npx skills add https://github.com/insforge/agent-skills --skill insforge

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。

用于数据库查询、身份验证、文件上传、AI 功能、实时消息传递和边缘功能调用的客户端 SDK。

六个核心模块:数据库(CRUD、过滤器、RPC)、身份验证(注册/登录、OAuth、会话)、存储(上传/下载/删除)、功能(边缘功能调用)、人工智能(聊天、图像生成、嵌入)和实时(WebSocket 通道、发布/订阅) 所有 SDK 方法返回 { data, error } 物体;数据库插入需要数组格式,存储操作都需要url 和钥匙 保存到数据库 实时通道使用SQL触发器自动发布数据库更改; RLS 策略通过 realtime.channels 控制订阅和发布访问 和实时消息 桌子 包括常见场景(订单访问、聊天成员资格)的 RLS 策略模式以及通道模式创建和触发器配置的最佳实践

来源仓库内路径:skills/insforge

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 Codex、Claude、Cursor、Gemini CLI;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 npx skills add https://github.com/insforge/agent-skills --skill insforge;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

insforge 怎么安装?

在终端执行 npx skills add https://github.com/insforge/agent-skills --skill insforge。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

insforge 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。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。

insforge 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。

insforge 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

insforge 用的人多吗?

公开数据是累计安装 155,976 次,近一周安装 6,740 次,下载 54,672 次,GitHub 13 星。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。

insforge 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

  • 同时适配 4 种工具,换客户端不用重新找替代品。

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。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 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 Codex,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

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 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
insforge

insforge 数据库 · 本页

155,97613接触敏感数据未明确有提醒

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

InsForge SDK Skill

This skill covers client-side SDK integration using @insforge/sdk. For backend infrastructure operations (creating tables, inspecting schema, deploying functions, secrets, managing storage buckets, configuring Stripe keys/catalog, website deployments, cron job and schedules, logs, etc.), use the insforge-cli skill.

Quick Setup

1. Install the SDK

npm install @insforge/sdk@latest

2. Set up environment variables

Before using the SDK, create a .env file (or .env.local for Next.js) in your project root with your InsForge URL and anon key.

How to get your URL and anon key
  1. Ensure the project is linked. Check for .insforge/project.json in the project root.

- If it doesn't exist, run npx @insforge/cli link (existing project) or npx @insforge/cli create (new project) to generate it.

  1. Get the anon key via the CLI: npx @insforge/cli secrets get ANON_KEY
  2. Get the URL from the oss_host field in .insforge/project.json (e.g., https://myapp.us-east.insforge.app).
  3. Write both values to the .env file using the correct framework prefix (see table below).

Use the correct environment variable prefix and access pattern for your framework:

Framework.env fileVariablesAccess Pattern
Next.js.env.localNEXT_PUBLIC_INSFORGE_URL, NEXT_PUBLIC_INSFORGE_ANON_KEYprocess.env.NEXT_PUBLIC_*
Vite (React, Vue, Svelte).envVITE_INSFORGE_URL, VITE_INSFORGE_ANON_KEYimport.meta.env.VITE_*
Astro.envPUBLIC_INSFORGE_URL, PUBLIC_INSFORGE_ANON_KEYimport.meta.env.PUBLIC_*
SvelteKit.envPUBLIC_INSFORGE_URL, PUBLIC_INSFORGE_ANON_KEYimport {env} from '$env/dynamic/public'
Create React App.envREACT_APP_INSFORGE_URL, REACT_APP_INSFORGE_ANON_KEYprocess.env.REACT_APP_*
Node.js / Server.envINSFORGE_URL, INSFORGE_ANON_KEYprocess.env.*

Example .env.local for Next.js:

NEXT_PUBLIC_INSFORGE_URL=https://your-appkey.us-east.insforge.app
NEXT_PUBLIC_INSFORGE_ANON_KEY=eyJhbGciOiJIUzI1NiIs...
Important: Never commit .env files to version control. Add .env, .env.local, and .env*.local to your .gitignore (keep .env.example for documenting required variables).

3. Initialize the client

import { createClient } from '@insforge/sdk'

// Next.js / CRA: use process.env
const insforge = createClient({
  baseUrl: process.env.NEXT_PUBLIC_INSFORGE_URL,
  anonKey: process.env.NEXT_PUBLIC_INSFORGE_ANON_KEY
})

// Vite / Astro: use import.meta.env
const insforge = createClient({
  baseUrl: import.meta.env.VITE_INSFORGE_URL,
  anonKey: import.meta.env.VITE_INSFORGE_ANON_KEY
})

Module Reference

What Each Module Covers

ModuleContent
DatabaseCRUD operations, filters, pagination, RPC calls
AuthSign up/in, OAuth, sessions, profiles, password reset
StorageUpload, download, delete files; write RLS policies for buckets
FunctionsInvoke edge functions
AIChat completions, image generation, embeddings
EmailSend custom transactional HTML emails (welcome, newsletter, notifications)
PaymentsStripe Checkout Sessions, subscriptions, and Billing Portal redirects
Real-timeConnect, subscribe, publish events, and track presence snapshots plus join/leave deltas

Guides

GuideWhen to Use
database/postgres-rls.mdWriting or reviewing RLS policies — covers infinite recursion prevention, SECURITY DEFINER patterns, performance tips, and common InsForge RLS patterns
storage/postgres-rls.mdWriting RLS policies for storage.objects — owner-only, public-read, path-scoped, team-shared, and the NULL uploaded_by caveat for mixed REST + S3 buckets
database/pgvector.mdBuilding semantic search, recommendations, or RAG — covers the vector extension, schema/dimensions, distance operators, HNSW/IVFFlat indexes, and RPC similarity search
ai/embeddings-and-rag.mdGenerating embeddings through the InsForge AI gateway, storing them in pgvector, and wiring up a basic RAG pipeline with chat completions
payments/backend-configuration.mdConfiguring Stripe keys, syncing catalog, creating products/prices, webhooks, and portal RLS before app integration

Real-time Configuration

For real-time channels and database triggers, use SQL migrations or database admin tooling to configure channels, triggers, and policies. The real-time SDK is for frontend event handling and messaging, not backend configuration.

Create Database Triggers

Automatically publish events when database records change.

-- Create trigger function
CREATE OR REPLACE FUNCTION notify_order_changes()
RETURNS TRIGGER AS $$
BEGIN
  PERFORM realtime.publish(
    'order:' || NEW.id::text,    -- channel
    TG_OP || '_order',           -- event: INSERT_order, UPDATE_order
    jsonb_build_object(
      'id', NEW.id,
      'status', NEW.status,
      'total', NEW.total
    )
  );
  RETURN NEW;
END;
$$ LANGUAGE plpgsql SECURITY DEFINER;

-- Attach to table
CREATE TRIGGER order_realtime
  AFTER INSERT OR UPDATE ON orders
  FOR EACH ROW
  EXECUTE FUNCTION notify_order_changes();
Conditional Trigger (Status Changes Only)
CREATE OR REPLACE FUNCTION notify_order_status()
RETURNS TRIGGER AS $$
BEGIN
  PERFORM realtime.publish(
    'order:' || NEW.id::text,
    'status_changed',
    jsonb_build_object('id', NEW.id, 'status', NEW.status)
  );
  RETURN NEW;
END;
$$ LANGUAGE plpgsql SECURITY DEFINER;

CREATE TRIGGER order_status_trigger
  AFTER UPDATE ON orders
  FOR EACH ROW
  WHEN (OLD.status IS DISTINCT FROM NEW.status)
  EXECUTE FUNCTION notify_order_status();
Access Control (RLS)

RLS is disabled by default. To restrict channel access:

  • Enable RLS
ALTER TABLE realtime.channels ENABLE ROW LEVEL SECURITY;
ALTER TABLE realtime.messages ENABLE ROW LEVEL SECURITY;
  • Restrict Subscribe (SELECT on channels)
CREATE POLICY "users_subscribe_own_orders"
ON realtime.channels FOR SELECT
TO authenticated
USING (
  pattern = 'order:%'
  AND EXISTS (
    SELECT 1 FROM orders
    WHERE id = NULLIF(split_part(realtime.channel_name(), ':', 2), '')::uuid
      AND user_id = auth.uid()
  )
);
  • Restrict Publish (INSERT on messages)
CREATE POLICY "members_publish_chat"
ON realtime.messages FOR INSERT
TO authenticated
WITH CHECK (
  channel_name LIKE 'chat:%'
  AND EXISTS (
    SELECT 1 FROM chat_members
    WHERE room_id = NULLIF(split_part(channel_name, ':', 2), '')::uuid
      AND user_id = auth.uid()
  )
);
  • Quick Reference
TaskSQL
Create channelINSERT INTO realtime.channels (pattern, description, enabled) VALUES (...)
Create triggerCREATE TRIGGER... EXECUTE FUNCTION...
Publish from SQLPERFORM realtime.publish(channel, event, payload)
Enable RLSALTER TABLE realtime.channels ENABLE ROW LEVEL SECURITY
Best Practices
  1. Create channel patterns first before subscribing from frontend

- Insert channel patterns into realtime.channels table - Ensure enabled is set to true

  1. Use specific channel patterns

- Use wildcard % patterns for dynamic channels (e.g., order:% for order:123) - Use exact patterns for global channels (e.g., notifications)

Common Mistakes
MistakeSolution
Subscribing to undefined channel patternCreate channel pattern in realtime.channels first
Channel not receiving messagesEnsure channel enabled is true
Publishing without triggerCreate database trigger to auto-publish on changes
Recommended Workflow
1. Create channel patterns   → INSERT INTO realtime.channels
2. Ensure enabled = true     → Set enabled to true
3. Create triggers if needed → Auto-publish on database changes
4. Proceed with SDK subscribe → Use channel name matching pattern

Backend Configuration (Not Yet in CLI)

These modules still require HTTP API calls because the CLI does not yet support them:

SDK Quick Reference

All SDK methods return {data, error}.

ModuleMethods
insforge.database.from().select(), .insert(), .update(), .delete(), .rpc()
insforge.auth.signUp(), .signInWithPassword(), .signInWithOAuth(), .signOut(), .getCurrentUser()
insforge.storage.from().upload(), .uploadAuto(), .download(), .remove()
insforge.functions.invoke()
insforge.ai.chat.completions.create(), .images.generate(), .embeddings.create()
insforge.realtime.connect(), .subscribe(), .publish(), .on(), .disconnect()
insforge.emails.send({to, subject, html, cc?, bcc?, from?, replyTo?})
insforge.payments.createCheckoutSession(), .createCustomerPortalSession()

Important Notes

  • Database inserts require array format: insert([{...}]) not insert({...})
  • Next.js / SSR auth: Use createClient({isServerMode: true}), keep tokens in httpOnly cookies, and perform auth flows on the server. See auth/sdk-integration.md
  • Storage: Save both url AND key to database for download/delete operations
  • Functions invoke URL: /functions/{slug} (without /api prefix)
  • Payments: Configure Stripe keys/catalog with npx @insforge/cli payments... first; frontend code only creates Checkout/Portal sessions.
  • Payment RLS: Before subscription checkout or Billing Portal UI, add app-specific RLS on payments.checkout_sessions and payments.customer_portal_sessions.
  • Use Tailwind CSS v3.4 (do not upgrade to v4)
  • Always local build before deploy: Prevents wasted build resources and faster debugging
  • Deprecated packages: @insforge/react, @insforge/nextjs, and @insforge/react-router are deprecated. Do NOT install or use them. Use @insforge/sdk directly for all features including authentication.
  • Deployment: Include a vercel.json in the project root for SPA routing (React, React Router apps). The download-template tool includes this automatically.

平台分布

Codex

33.86%
按下载量换算18,512

Claude

29.79%
按下载量换算16,287

Cursor

19%
按下载量换算10,388

Gemini CLI

9.82%
按下载量换算5,369

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。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