convex-actions-general
About
This skill provides comprehensive guidance for implementing Convex actions, HTTP endpoints, and database schemas using TypeScript. Use it when working on external API calls, webhooks, validators, file storage, and environment variables. It covers essential patterns, function definitions, and advanced features for Convex development.
Documentation
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.tswith 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()andv.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
apiandinternalobjects - 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.tsstructure- System fields (
_id,_creationTime) - Table definitions with validators
- System fields (
- 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()andcrons.cron() - Scheduler: Using
ctx.scheduler.runAfter()from mutations/actions - Auth state propagation (doesn't propagate to scheduled jobs)
- Timing constraints and best practices
- Crons: Using
File Storage
- Upload URL generation with
ctx.storage.generateUploadUrl() - Signed URL retrieval with
ctx.storage.getUrl() - File metadata from
_storagesystem 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 IDsDoc<'tableName'>types for document type safetyRecord<KeyType, ValueType>with proper typingas constfor discriminated unions- Type annotations for same-file function calls
@types/nodefor Node.js modules
How to Use This Skill
- Read the reference documentation at
references/actions-and-general.mdfor comprehensive patterns - Follow the syntax for defining actions with proper Node.js module handling
- Use validators correctly for all function arguments and schema fields
- Design schemas with appropriate indexes for your access patterns
- Set up environment variables for secrets and configuration
- Implement scheduling for background tasks using crons or the scheduler
- Handle file storage with proper URL generation and metadata lookup
- Understand limits and design applications to respect them
- Use TypeScript strictly with
Idtypes 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.
Quick Install
/plugin add https://github.com/Sstobo/convex-skills/tree/main/convex-actions-generalCopy and paste this command in Claude Code to install this skill
GitHub 仓库
Related Skills
evaluating-llms-harness
TestingThis Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
Algorithmic Art Generation
MetaThis skill helps developers create algorithmic art using p5.js, focusing on generative art, computational aesthetics, and interactive visualizations. It automatically activates for topics like "generative art" or "p5.js visualization" and guides you through creating unique algorithms with features like seeded randomness, flow fields, and particle systems. Use it when you need to build reproducible, code-driven artistic patterns.
webapp-testing
TestingThis Claude Skill provides a Playwright-based toolkit for testing local web applications through Python scripts. It enables frontend verification, UI debugging, screenshot capture, and log viewing while managing server lifecycles. Use it for browser automation tasks but run scripts directly rather than reading their source code to avoid context pollution.
