agentleFS
waynesuttonwaynesutton/markdown-site/.claude/function-creator/SKILL.md

function-creator

Skill628 starsChanged 7 months agoMITView on GitHub

Create Convex queries, mutations, and actions with proper validation, authentication, and error handling. Use when implementing new API endpoints.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run in ACID transactions

    line 47
  2. Very specific:

    Note: Keep queries and mutations in files without "use node".

    line 114
  3. Distinctive:

    Always: verify auth in public functions:

    line 128
  4. Distinctive:

    Always: verify ownership/permissions:

    line 137

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Convex Function Creator

Generate secure, type-safe Convex functions following all best practices.

When to Use

  • Creating new query functions (read data)
  • Creating new mutation functions (write data)
  • Creating new action functions (external APIs, long-running)
  • Adding API endpoints to your Convex backend

Function Types

Queries (Read-Only)

  • Can only read from database
  • Cannot modify data or call external APIs
  • Cached and reactive
  • Run in transactions
import { query } from "./_generated/server";
import { v } from "convex/values";

export const getTask = query({
  args: { taskId: v.id("tasks") },
  returns: v.union(v.object({
    _id: v.id("tasks"),
    text: v.string(),
    completed: v.boolean(),
  }), v.null()),
  handler: async (ctx, args) => {
    return await ctx.db.get(args.taskId);
  },
});

Mutations (Transactional Writes)

  • Can read and write to database
  • Cannot call external APIs
  • Run in ACID transactions
import { mutation } from "./_generated/server";
import { v } from "convex/values";

export const createTask = mutation({
  args: {
    text: v.string(),
    priority: v.optional(v.union(
      v.literal("low"),
      v.literal("medium"),
      v.literal("high")
    )),
  },
  returns: v.id("tasks"),
  handler: async (ctx, args) => {
    const identity = await ctx.auth.getUserIdentity();
    if (!identity) throw new Error("Not authenticated");

    return await ctx.db.insert("tasks", {
      text: args.text,
      priority: args.priority ?? "medium",
      completed: false,
      createdAt: Date.now(),
    });
  },
});

Actions (External + Non-Transactional)

  • Can call external APIs (fetch, AI, etc.)
  • Can call mutations via ctx.runMutation
  • Cannot directly access database
  • Use "use node" directive when needing Node.js APIs
"use node";

import { action } from "./_generated/server";
import { api } from "./_generated/api";
import { v } from "convex/values";
import OpenAI from "openai";

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

export const generateTaskSuggestion = action({
  args: { prompt: v.string() },
  returns: v.string(),
  handler: async (ctx, args) => {
    const identity = await ctx.auth.getUserIdentity();
    if (!identity) throw new Error("Not authenticated");

    const completion = await openai.chat.completions.create({
      model: "gpt-4",
      messages: [{ role: "user", content: args.prompt }],
    });

    const suggestion = completion.choices[0].message.content;

    await ctx.runMutation(api.tasks.createTask, { text: suggestion });
    return suggestion;
  },
});

Note: Keep queries and mutations in files without "use node". Actions that need Node.js go in separate files with "use node".

Required Components

1. Argument Validation

Always define args with validators.

2. Return Type Validation

Always define returns.

3. Authentication Check

Always verify auth in public functions:

const identity = await ctx.auth.getUserIdentity();
if (!identity) throw new Error("Not authenticated");

4. Authorization Check

Always verify ownership/permissions:

const task = await ctx.db.get(args.taskId);
if (!task) throw new Error("Task not found");
if (task.userId !== user._id) throw new Error("Unauthorized");

Internal Functions

For backend-only functions (called by scheduler, other functions):

import { internalMutation } from "./_generated/server";

export const processExpiredTasks = internalMutation({
  args: {},
  handler: async (ctx) => {
    const now = Date.now();
    const expired = await ctx.db
      .query("tasks")
      .withIndex("by_due_date", q => q.lt("dueDate", now))
      .collect();

    for (const task of expired) {
      await ctx.db.patch(task._id, { status: "expired" });
    }
  },
});

Checklist

  • args defined with validators
  • returns defined with validator
  • Authentication check (ctx.auth.getUserIdentity())
  • Authorization check (ownership/permissions)
  • All promises awaited
  • Indexed queries (no .filter() on queries)
  • Error handling with descriptive messages
  • Scheduled functions use internal.* not api.*
  • If using Node.js APIs: "use node" at top of file
  • If file has "use node": Only actions (no queries/mutations)

Source: https://github.com/get-convex/convex-agent-plugins

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.