Token导航 LogoToken导航

Agent Skill · 研究检索

convex-actions-general(凸行动一般)

convex-actions-general 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 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。

convex-actions-general 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 52,984。

总安装 541周安装 23GitHub Stars 23下载 190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:convex-actions-general(凸行动一般)
来源仓库:https://github.com/sstobo/convex-skills
仓库路径:skills/convex-actions-general
安装命令:
npx skills add https://github.com/sstobo/convex-skills --skill convex-actions-general
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/sstobo/convex-skills --skill convex-actions-general

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。

convex-actions-general 提供 Convex 动作、HTTP 端点、校验器与数据库 schema 设计指南。适用于实现外部 API 调用、长期任务或公共接口开发的场景。覆盖环境变量管理、文件存储、调度配置与 TypeScript 最佳实践。

严禁使用 any 类型,否则会导致构建失败,需遵循严格类型约束。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/convex-actions-general

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/sstobo/convex-skills --skill convex-actions-ge…;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

convex-actions-general 怎么安装?

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

convex-actions-general 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。

convex-actions-general 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。

convex-actions-general 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。

convex-actions-general 用的人多吗?

公开数据是累计安装 541 次,近一周安装 23 次,下载 190 次,GitHub 23 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。

convex-actions-general 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
convex-actions-general

凸行动一般 · 本页

54123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

Convex Actions and General Guidelines Skill

This skill provides comprehensive guidance for Convex actions, HTTP endpoints, validators, schema design, file storage, environment variables, scheduling, and TypeScript best practices.

When to Use This Skill

Use this skill when:

  • Implementing action functions for external API calls and long-running tasks
  • Creating HTTP endpoints for webhooks or public APIs
  • Defining validators for function arguments and database schemas
  • Designing database schemas with tables, indexes, and search capabilities
  • Setting up environment variables for secrets and configuration
  • Implementing cron jobs and scheduled tasks
  • Working with file storage for uploads and downloads
  • Using Convex-specific TypeScript patterns and types
  • Understanding Convex limits and performance constraints

Skill Resources

This skill includes comprehensive reference documentation in references/actions-and-general.md that covers:

Actions and HTTP

  • Actions: Defining actions with "use node" for Node.js modules

- V8 vs Node runtime differences - Action limitations (10-minute timeout, no database access) - Calling external APIs and services

  • HTTP Endpoints: Setting up convex/http.ts with httpRouter

- Path registration and exact matching - Request/response handling - Method definitions (POST, GET, etc.)

Validators and Types

  • Complete validator reference for all Convex types
  • Common validators: v.object(), v.array(), v.string(), v.number(), v.id(), v.boolean(), v.null()
  • Discriminated union types with v.union() and v.literal()
  • ASCII field name requirements for objects
  • Size and element count limits

Function Development

  • New function syntax for all function types
  • Function registration patterns (query, mutation, action, internalQuery, internalMutation, internalAction)
  • Function calling patterns across runtimes
  • Function references via api and internal objects
  • File-based routing conventions

API Design

  • Organizing public and private functions
  • Thoughtful file structure within convex/ directory
  • Public vs. internal function visibility
  • API surface consistency

Database and Schema

  • Schema Definition: convex/schema.ts structure

- System fields (_id, _creationTime) - Table definitions with validators

  • Indexes: Creating efficient indexes

- Built-in indexes (by_id, by_creation_time) - Custom index naming and field ordering - Multiple field indexes for complex queries

  • Full Text Search: Search index definitions

- Search fields and filter fields - Nested field paths with dot notation

Environment and Configuration

  • Environment Variables: Using process.env

- Storing secrets (API keys, credentials) - Per-deployment configuration - Access from any function type

  • Scheduling:

- Crons: Using crons.interval() and crons.cron() - Scheduler: Using ctx.scheduler.runAfter() from mutations/actions - Auth state propagation (doesn't propagate to scheduled jobs) - Timing constraints and best practices

File Storage

  • Upload URL generation with ctx.storage.generateUploadUrl()
  • Signed URL retrieval with ctx.storage.getUrl()
  • File metadata from _storage system table
  • Blob conversion for storage operations
  • Complete example: image upload in chat application

Limits and Performance

  • Function arguments and return values: 8 MiB maximum
  • Database operations: 8192 document writes per mutation, 16384 reads per query
  • Execution timeouts: 1 second for queries/mutations, 10 minutes for actions
  • Array element limits: 8192 maximum
  • Object/Record field limits: 1024 maximum
  • Nesting depth: 16 maximum
  • Record size: 1 MiB maximum
  • HTTP streaming: 20 MiB maximum output

TypeScript

  • Id<'tableName'> types for strict document IDs
  • Doc<'tableName'> types for document type safety
  • Record<KeyType, ValueType> with proper typing
  • as const for discriminated unions
  • Type annotations for same-file function calls
  • @types/node for Node.js modules

How to Use This Skill

  1. Read the reference documentation at references/actions-and-general.md for comprehensive patterns
  2. Follow the syntax for defining actions with proper Node.js module handling
  3. Use validators correctly for all function arguments and schema fields
  4. Design schemas with appropriate indexes for your access patterns
  5. Set up environment variables for secrets and configuration
  6. Implement scheduling for background tasks using crons or the scheduler
  7. Handle file storage with proper URL generation and metadata lookup
  8. Understand limits and design applications to respect them
  9. Use TypeScript strictly with Id types and proper generics

Key General Guidelines

  • ALWAYS use argument validators for all functions (queries, mutations, actions)
  • Do NOT store file URLs in the database; store file IDs instead
  • Remapping non-ASCII characters (emoji) to ASCII codes before storing in objects
  • Auth state does NOT propagate to scheduled jobs; use internal functions
  • Scheduled functions should not run more than once every 10 seconds
  • Never call actions from other actions unless crossing runtimes (V8 to Node)
  • Objects in Convex must have ASCII-only field names
  • Be strict with TypeScript types, especially for document IDs

Example: Complete Action with HTTP Endpoint

// convex/ai.ts
"use node";
import { action } from "./_generated/server";
import { v } from "convex/values";
import { internal } from "./_generated/api";
import OpenAI from "openai";

const openai = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
});

export const generateResponse = action({
  args: {
    channelId: v.id("channels"),
  },
  handler: async (ctx, args) => {
    // Actions can't access ctx.db, but can call mutations
    const context = await ctx.runQuery(internal.functions.loadContext, {
      channelId: args.channelId,
    });

    const response = await openai.chat.completions.create({
      model: "gpt-4o-mini",
      messages: context,
    });

    const content = response.choices[0].message.content;
    if (!content) throw new Error("No content in response");

    await ctx.runMutation(internal.functions.writeAgentResponse, {
      channelId: args.channelId,
      content,
    });

    return null;
  },
});

Example: HTTP Endpoint

// convex/http.ts
import { httpRouter } from "convex/server";
import { httpAction } from "./_generated/server";

const http = httpRouter();

http.route({
  path: "/webhook",
  method: "POST",
  handler: httpAction(async (ctx, req) => {
    const body = await req.json();
    // Process webhook payload
    return new Response(JSON.stringify({ success: true }), { status: 200 });
  }),
});

export default http;

For more detailed information and additional patterns, refer to the complete reference documentation.

平台分布

Codex

37.19%
按下载量换算71

Claude

32.31%
按下载量换算61

Cursor

18.63%
按下载量换算35

Gemini CLI

9.7%
按下载量换算18

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