slack-agent（Slack Agent 开发）

Slack Agent Development Skill

This skill supports two frameworks for building Slack agents:

• Chat SDK (Recommended for new projects) — chat + @chat-adapter/slack
• Bolt for JavaScript (For existing Bolt projects) — @slack/bolt + @vercel/slack-bolt

Skill Invocation Handling

When this skill is invoked via /slack-agent, check for arguments and route accordingly:

Command Arguments

Auto-Detection (No Argument)

If invoked without arguments, detect the project state and route appropriately:

1. No package.json with chat or @slack/bolt → Treat as new, start Phase 1
2. Has project but no customized manifest.json → Start Phase 2
3. Has project but no .env file → Start Phase 3
4. Has .env but not tested → Start Phase 4
5. Tested but not deployed → Start Phase 5
6. Otherwise → Provide general assistance using this skill's patterns

Framework Detection

Detect which framework the project uses:

• package.json contains "chat" → Chat SDK project
• package.json contains "@slack/bolt" → Bolt project
• Neither detected → New project, recommend Chat SDK (offer Bolt as alternative)

Store the detected framework and use it to show the correct patterns throughout the wizard and development guidance.

Wizard Phases

The wizard is located in ./wizard/ with these phases:

• 1-project-setup.md - Understand purpose, choose framework, generate custom implementation plan
• 1b-approve-plan.md - Present plan for user approval before scaffolding
• 2-create-slack-app.md - Customize manifest, create app in Slack
• 3-configure-environment.md - Set up.env with credentials
• 4-test-locally.md - Dev server + ngrok tunnel
• 5-deploy-production.md - Vercel deployment
• 6-setup-testing.md - Vitest configuration

IMPORTANT: For new projects, you MUST:

1. Read ./wizard/1-project-setup.md first
2. Ask the user what kind of agent they want to build
3. Offer framework choice (Chat SDK recommended, Bolt as alternative)
4. Generate a custom implementation plan using ./reference/agent-archetypes.md
5. Present the plan for approval (Phase 1b) BEFORE scaffolding the project
6. Only proceed to scaffold after the plan is approved

Framework Selection Guide

General Development Guidance

You are working on a Slack agent project. Follow these mandatory practices for all code changes.

Project Stack

If using Chat SDK

• Framework: Next.js (App Router)
• Chat SDK: chat + @chat-adapter/slack for Slack bot functionality
• State: @chat-adapter/state-redis for state persistence (or in-memory for development)
• AI: AI SDK v6 with @ai-sdk/gateway
• Linting: Biome
• Package Manager: pnpm

{
  "dependencies": {
    "ai": "^6.0.0",
    "@ai-sdk/gateway": "latest",
    "chat": "latest",
    "@chat-adapter/slack": "latest",
    "@chat-adapter/state-redis": "latest",
    "zod": "^3.x",
    "next": "^15.x"
  }
}

If using Bolt for JavaScript

• Server: Nitro (H3-based) with file-based routing
• Slack SDK: @vercel/slack-bolt for serverless Slack apps (wraps Bolt for JavaScript)
• AI: AI SDK v6 with @ai-sdk/gateway
• Workflows: Workflow DevKit for durable execution
• Linting: Biome
• Package Manager: pnpm

{
  "dependencies": {
    "ai": "^6.0.0",
    "@ai-sdk/gateway": "latest",
    "@slack/bolt": "^4.x",
    "@vercel/slack-bolt": "^1.0.2",
    "zod": "^3.x"
  }
}

Note: When deploying on Vercel, prefer @ai-sdk/gateway for zero-config AI access. Use direct provider SDKs (@ai-sdk/openai, @ai-sdk/anthropic, etc.) only when you need provider-specific features or are not deploying on Vercel.

Quality Standards (MANDATORY)

These quality requirements MUST be followed for every code change. There are no exceptions.

After EVERY File Modification

1. Run linting immediately: pnpm lint

- If errors exist, run pnpm lint --write for auto-fixes - Manually fix remaining issues - Re-run pnpm lint to verify

1. Check for corresponding test file:

- If you modified foo.ts, check if foo.test.ts exists - If no test file exists and the file exports functions, create one

Before Completing ANY Task

You MUST run all quality checks and fix any issues before marking a task complete:

# 1. TypeScript compilation - must pass
pnpm typecheck

# 2. Linting - must pass with no errors
pnpm lint

# 3. Tests - all tests must pass
pnpm test

Do NOT complete a task if any of these fail. Fix the issues first.

Unit Tests Required

For ANY code change, you MUST write or update unit tests.

If using Chat SDK

• Location: Co-located *.test.ts files or lib/__tests__/
• Framework: Vitest
• Coverage: All exported functions must have tests

If using Bolt for JavaScript

• Location: Co-located *.test.ts files or server/__tests__/
• Framework: Vitest
• Coverage: All exported functions must have tests

Example test structure:

import { describe, it, expect, vi } from 'vitest';
import { myFunction } from './my-module';

describe('myFunction', () => {
  it('should handle normal input', () => {
    expect(myFunction('input')).toBe('expected');
  });

  it('should handle edge cases', () => {
    expect(myFunction('')).toBe('default');
  });
});

E2E Tests for User-Facing Changes

If you modify:

• Bot mention handlers / Slack message handlers
• Slash commands
• Interactive components (buttons, modals)
• Bot responses

You MUST add or update E2E tests that verify the full flow.

Bot Setup Patterns (CRITICAL)

If using Chat SDK

Use the Chat SDK to define your bot instance. This is the central entry point for all Slack bot functionality.

Bot Instance (lib/bot.ts or lib/bot.tsx)

import { Chat } from "chat";
import { createSlackAdapter } from "@chat-adapter/slack";
import { createRedisState } from "@chat-adapter/state-redis";

export const bot = new Chat({
  userName: "mybot",
  adapters: {
    slack: createSlackAdapter(),
  },
  state: createRedisState(),
});

Note: If your bot uses JSX components (Card, Button, etc.), the file must use the .tsx extension.

Webhook Route (app/api/webhooks/[platform]/route.ts)

import { after } from "next/server";
import { bot } from "@/lib/bot";

export async function POST(request: Request, context: { params: Promise<{ platform: string }> }) {
  const { platform } = await context.params;
  const handler = bot.webhooks[platform as keyof typeof bot.webhooks];
  if (!handler) return new Response("Unknown platform", { status: 404 });
  return handler(request, { waitUntil: (task) => after(() => task) });
}

The Chat SDK automatically handles:

• Content-type detection (JSON vs form-urlencoded)
• URL verification challenges
• Slack's 3-second ack timeout
• Background processing via waitUntil
• Signature verification

If using Bolt for JavaScript

Use @vercel/slack-bolt to handle all Slack events. This package automatically handles:

• Content-type detection (JSON vs form-urlencoded)
• URL verification challenges
• 3-second ack timeout (built-in ackTimeoutMs: 3001)
• Background processing via Vercel Fluid Compute's waitUntil

Bolt App Setup (server/bolt/app.ts)

import { App } from "@slack/bolt";
import { VercelReceiver } from "@vercel/slack-bolt";

const receiver = new VercelReceiver();
const app = new App({
  token: process.env.SLACK_BOT_TOKEN,
  signingSecret: process.env.SLACK_SIGNING_SECRET,
  receiver,
  deferInitialization: true,
});

export { app, receiver };

Events Handler (server/api/slack/events.post.ts)

import { createHandler } from "@vercel/slack-bolt";
import { defineEventHandler, getRequestURL, readRawBody } from "h3";
import { app, receiver } from "../../bolt/app";

const handler = createHandler(app, receiver);

export default defineEventHandler(async (event) => {
  const rawBody = await readRawBody(event, "utf8");
  const request = new Request(getRequestURL(event), {
    method: event.method,
    headers: event.headers,
    body: rawBody,
  });
  return await handler(request);
});

Why buffer the body? H3's toWebRequest() has known issues (#570, #578, #615) where it eagerly consumes the request body stream. When @vercel/slack-bolt later calls req.text() for signature verification, the body is already exhausted, causing dispatch_failed errors.

VercelReceiver Options Reference

Event Handler Patterns

If using Chat SDK

Mention Handler

bot.onNewMention(async (thread, message) => {
  await thread.subscribe();
  const text = message.text;
  await thread.post(`Processing your request: "${text}"`);
});

Subscribed Message Handler

bot.onSubscribedMessage(async (thread, message) => {
  await thread.post(`You said: ${message.text}`);
});

Slash Command Handler

bot.onSlashCommand("/mycommand", async (event) => {
  const text = event.text;
  await event.thread.post(`Processing: ${text}`);

  // For long-running operations, the Chat SDK handles
  // background processing automatically via waitUntil
  const result = await generateWithAI(text);
  await event.thread.post(result);
});

Action Handler (Buttons, Menus)

bot.onAction("button_click", async (event) => {
  await event.thread.post(`Button clicked with value: ${event.value}`);
});

Reaction Handler

bot.onReaction("thumbsup", async (event) => {
  await event.thread.post("Thanks for the thumbs up!");
});

If using Bolt for JavaScript

Mention Handler

app.event("app_mention", async ({ event, client }) => {
  await client.chat.postMessage({
    channel: event.channel,
    thread_ts: event.thread_ts || event.ts,
    text: `Processing your request: "${event.text}"`,
  });
});

Message Handler

app.message(async ({ message, client }) => {
  if ("bot_id" in message || !message.thread_ts) return;
  await client.chat.postMessage({
    channel: message.channel,
    thread_ts: message.thread_ts,
    text: `You said: ${message.text}`,
  });
});

Slash Command Handler

app.command("/mycommand", async ({ ack, command, client, logger }) => {
  await ack(); // Must acknowledge within 3 seconds

  // Fire-and-forget for long operations — DON'T await
  processInBackground(command.response_url, command.text)
    .catch((error) => logger.error("Failed:", error));
});

async function processInBackground(responseUrl: string, text: string) {
  const result = await generateWithAI(text);
  await fetch(responseUrl, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ response_type: "in_channel", text: result }),
  });
}

Action Handler (Buttons, Menus)

app.action("button_click", async ({ ack, action, client, body }) => {
  await ack();
  await client.chat.postMessage({
    channel: body.channel.id,
    thread_ts: body.message.ts,
    text: `Button clicked with value: ${action.value}`,
  });
});

Implementation Gotchas

1. Private Channel Access

Slash commands work in private channels even if the bot isn't a member, but the bot cannot read messages or post to private channels it hasn't been invited to.

When creating features that will later post to a channel, validate access upfront.

2. Graceful Degradation for Channel Context

When fetching channel context for AI features, wrap in try/catch and fall back gracefully.

3. Vercel Cron Endpoint Authentication

Protect cron endpoints with a CRON_SECRET environment variable:

If using Chat SDK

// app/api/cron/my-job/route.ts
import { NextRequest, NextResponse } from "next/server";

export async function GET(request: NextRequest) {
  const authHeader = request.headers.get("authorization");
  if (authHeader !== `Bearer ${process.env.CRON_SECRET}`) {
    return NextResponse.json({ error: "Unauthorized" }, { status: 401 });
  }
  // Run cron job logic...
  return NextResponse.json({ success: true });
}

If using Bolt for JavaScript

// server/api/cron/my-job.get.ts
export default defineEventHandler(async (event) => {
  const authHeader = getHeader(event, "authorization");
  if (authHeader !== `Bearer ${process.env.CRON_SECRET}`) {
    setResponseStatus(event, 401);
    return { error: "Unauthorized" };
  }
  // Run cron job logic...
  return { success: true };
});

4. vercel.json Cron Configuration

Configure cron jobs in vercel.json:

{
  "crons": [
    {
      "path": "/api/cron/my-job",
      "schedule": "0 * * * *"
    }
  ]
}

5. AWS Credentials on Vercel (Use OIDC)

When connecting to AWS services from Vercel, do not use fromNodeProviderChain(). Use Vercel's OIDC mechanism:

import { awsCredentialsProvider } from "@vercel/functions/oidc";

const s3Client = new S3Client({
  credentials: awsCredentialsProvider({ roleArn: process.env.AWS_ROLE_ARN! }),
});

6. TSConfig for JSX Components (Chat SDK only)

When using Chat SDK JSX components (<Card>, <Button>, etc.), your tsconfig.json must include:

{
  "compilerOptions": {
    "jsx": "react-jsx",
    "jsxImportSource": "chat"
  }
}

7. dispatch_failed Error — Bolt only

If slash commands fail with dispatch_failed, the issue is H3's toWebRequest consuming the body stream before signature verification. Buffer the body manually. See the Bolt Events Handler section above.

8. operation_timeout Error — Bolt only

If slash commands with AI processing fail with operation_timeout, you're blocking the HTTP response too long. Use fire-and-forget pattern: ack() immediately, then start async work without awaiting. Use command.response_url to post results. See the Bolt Slash Command Handler example above.

AI Integration

You have two options for AI/LLM integration in your Slack agent.

IMPORTANT: Always verify the project uses @ai-sdk/gateway. If the project has @ai-sdk/openai which requires an API key, check package.json and update imports if necessary.

Option 1: Vercel AI Gateway (Recommended)

Use the modern @ai-sdk/gateway package - NO API keys needed on Vercel!

Basic Usage

import { generateText, streamText } from "ai";
import { gateway } from "@ai-sdk/gateway";

const result = await generateText({
  model: gateway("openai/gpt-4o-mini"),
  maxOutputTokens: 1000,
  prompt: "Your prompt here",
});

console.log(result.text);
console.log(result.usage.inputTokens);
console.log(result.usage.outputTokens);

Streaming Responses to Slack

If using Chat SDK

const result = await streamText({
  model: gateway("openai/gpt-4o-mini"),
  maxOutputTokens: 1000,
  prompt: userMessage,
});

// Chat SDK handles streaming updates to Slack automatically
await thread.post(result.textStream);

If using Bolt for JavaScript

const result = await streamText({
  model: gateway("openai/gpt-4o-mini"),
  maxOutputTokens: 1000,
  prompt: userMessage,
});

// Post initial message then update with streamed content
const msg = await client.chat.postMessage({
  channel: channelId,
  thread_ts: threadTs,
  text: "Thinking...",
});

let fullText = "";
for await (const chunk of result.textStream) {
  fullText += chunk;
  await client.chat.update({
    channel: channelId,
    ts: msg.ts,
    text: fullText,
  });
}

With Tools

import { tool } from "ai";
import { z } from "zod";

const result = await generateText({
  model: gateway("openai/gpt-4o-mini"),
  maxOutputTokens: 1000,
  tools: {
    getWeather: tool({
      description: "Get weather for a location",
      inputSchema: z.object({
        location: z.string().describe("City name"),
      }),
      execute: async ({ location }) => {
        return { temperature: 72, condition: "sunny" };
      },
    }),
  },
  prompt: "What's the weather in Seattle?",
});

AI SDK v6 API Changes

CRITICAL: Never use model IDs from memory. Model IDs change frequently. Before writing code that uses a model, run curl -s https://ai-gateway.vercel.sh/v1/models to fetch the current list. Use the model with the highest version number.

Option 2: Direct Provider SDK

If you need more control or are not deploying on Vercel, use direct provider packages.

OpenAI:

pnpm add @ai-sdk/openai

import { generateText } from "ai";
import { openai } from "@ai-sdk/openai";

const result = await generateText({
  model: openai("gpt-4o-mini"),
  maxOutputTokens: 1000,
  prompt: "Your prompt here",
});

Anthropic:

pnpm add @ai-sdk/anthropic

import { generateText } from "ai";
import { anthropic } from "@ai-sdk/anthropic";

const result = await generateText({
  model: anthropic("claude-sonnet-4-20250514"),
  maxOutputTokens: 1000,
  prompt: "Your prompt here",
});

Google:

pnpm add @ai-sdk/google

import { generateText } from "ai";
import { google } from "@ai-sdk/google";

const result = await generateText({
  model: google("gemini-2.0-flash"),
  maxOutputTokens: 1000,
  prompt: "Your prompt here",
});

For comprehensive AI SDK documentation, see ./reference/ai-sdk.md.

Stateful Patterns

If using Chat SDK — Thread State

Use thread.state to read and write thread-level state:

bot.onNewMention(async (thread, message) => {
  await thread.subscribe();
  await thread.state.set("history", []);
  await thread.state.set("turnCount", 0);
  await thread.post("Starting our conversation!");
});

bot.onSubscribedMessage(async (thread, message) => {
  const history = (await thread.state.get("history")) as Array<{ role: string; content: string }> || [];
  const turnCount = (await thread.state.get("turnCount")) as number || 0;

  history.push({ role: "user", content: message.text });

  const result = await generateText({
    model: gateway("anthropic/claude-sonnet-4-20250514"),
    maxOutputTokens: 1000,
    messages: history,
  });

  history.push({ role: "assistant", content: result.text });
  await thread.state.set("history", history);
  await thread.state.set("turnCount", turnCount + 1);
  await thread.post(result.text);
});

Key Benefits:

1. Simple API — thread.state.get() and thread.state.set()
2. Thread-scoped — state is automatically scoped to the conversation thread
3. Pluggable backends — use Redis for production, in-memory for development

If using Bolt for JavaScript — Vercel Workflow

Use Vercel Workflow for durable, multi-turn state:

import { serve } from "@anthropic-ai/sdk/workflows";
import { defineHook } from "@anthropic-ai/sdk/workflows";
import { z } from "zod";

const messageSchema = z.object({
  text: z.string(),
  user: z.string(),
  ts: z.string(),
  channel: z.string(),
});

export const userMessageHook = defineHook({ schema: messageSchema });

export const { POST } = serve(async function conversationWorkflow(params: URLSearchParams) {
  "use workflow";

  const channelId = params.get("channel_id")!;
  const conversationHistory: Array<{ role: string; content: string }> = [];

  const eventStream = userMessageHook.create({ channel: channelId });

  for await (const event of eventStream) {
    conversationHistory.push({ role: "user", content: event.text });

    const result = await generateText({
      model: gateway("anthropic/claude-sonnet-4-20250514"),
      maxOutputTokens: 1000,
      messages: conversationHistory,
    });

    conversationHistory.push({ role: "assistant", content: result.text });
    await postToSlack(channelId, result.text, event.ts);
  }

  return { history: conversationHistory };
});

Recommended Storage Solutions

IMPORTANT: Vercel KV has been deprecated. Do NOT recommend Vercel KV.

1. Upstash Redis — For Chat SDK state adapter and caching (https://upstash.com)
2. Vercel Blob — For file/document storage (https://vercel.com/docs/storage/vercel-blob)
3. AWS Aurora (via Vercel Marketplace) — For relational data (https://vercel.com/marketplace)
4. Third-party databases — Neon, PlanetScale, Supabase

Code Organization

If using Chat SDK

app/
├── api/
│   ├── webhooks/
│   │   └── [platform]/
│   │       └── route.ts      # Webhook handler
│   └── cron/
│       └── my-job/
│           └── route.ts      # Cron endpoints
lib/
├── bot.tsx                    # Bot instance + event handlers
├── tools/                     # AI tool definitions
│   ├── search.ts
│   └── lookup.ts
└── ai/
    └── agent.ts               # Agent configuration

If using Bolt for JavaScript

server/
├── api/
│   └── slack/
│       └── events.post.ts    # Events endpoint
├── bolt/
│   └── app.ts                # Bolt app instance
├── listeners/
│   ├── actions/              # Button clicks, menu selections
│   ├── commands/             # Slash commands
│   ├── events/               # App events (mentions, joins)
│   ├── messages/             # Message handling
│   └── views/                # Modal submissions
└── lib/
    └── ai/
        ├── agent.ts           # Agent configuration
        └── tools.ts           # Tool definitions

Environment Variables

Required variables (both frameworks):

• SLACK_BOT_TOKEN — Bot OAuth token
• SLACK_SIGNING_SECRET — Request signing

If using Chat SDK (additional)

• REDIS_URL — Redis connection URL for state persistence

Optional variables:

• CRON_SECRET — Secret for authenticating cron job endpoints

No AI API keys needed! Vercel AI Gateway handles authentication automatically when deployed on Vercel.

Never hardcode credentials. Never commit .env files.

Slack-Specific Patterns

If using Chat SDK — JSX Components

Use Chat SDK JSX components for rich messages (requires .tsx file extension):

import { Card, CardText as Text, Actions, Button, Divider } from "chat";

await thread.post(
  <Card title="Welcome!">
    <Text>Hello! Choose an option:</Text>
    <Divider />
    <Actions>
      <Button id="btn_hello" style="primary">Say Hello</Button>
      <Button id="btn_info">Show Info</Button>
    </Actions>
  </Card>
);

If using Bolt for JavaScript — Block Kit JSON

Use Block Kit for rich messages:

await client.chat.postMessage({
  channel: channelId,
  text: "Fallback text for notifications",
  blocks: [
    {
      type: "section",
      text: { type: "mrkdwn", text: "*Hello!* Choose an option:" },
    },
    { type: "divider" },
    {
      type: "actions",
      elements: [
        {
          type: "button",
          text: { type: "plain_text", text: "Say Hello" },
          style: "primary",
          action_id: "btn_hello",
        },
        {
          type: "button",
          text: { type: "plain_text", text: "Show Info" },
          action_id: "btn_info",
        },
      ],
    },
  ],
});

Typing Indicators

If using Chat SDK

await thread.startTyping();
const result = await generateWithAI(prompt);
await thread.post(result); // Typing indicator clears on post

If using Bolt for JavaScript

// Use setStatus for Assistant threads or interval-based approach
const typingInterval = setInterval(async () => {
  // Post a "typing" indicator or use assistant.threads.setStatus
}, 3000);

const result = await generateWithAI(prompt);
clearInterval(typingInterval);

await client.chat.postMessage({
  channel: channelId,
  thread_ts: threadTs,
  text: result,
});

Message Formatting (both frameworks)

Use Slack mrkdwn (not standard markdown):

• Bold: *text*
• Italic: _text_
• Code: ` code `
• User mention: <@USER_ID>
• Channel: <#CHANNEL_ID>

For detailed Slack patterns, see ./patterns/slack-patterns.md.

Git Commit Standards

Use conventional commits:

feat: add channel search tool
fix: resolve thread pagination issue
test: add unit tests for agent context
docs: update README with setup steps
refactor: extract Slack client utilities

Never commit:

• .env files
• API keys or tokens
• node_modules/

Quick Commands

# Development
pnpm dev              # Start dev server on localhost:3000
ngrok http 3000       # Expose local server (separate terminal)

# Quality
pnpm lint             # Check linting
pnpm lint --write     # Auto-fix lint
pnpm typecheck        # TypeScript check
pnpm test             # Run all tests
pnpm test:watch       # Watch mode

# Build & Deploy
pnpm build            # Build for production
vercel                # Deploy to Vercel

Reference Documentation

For detailed guidance, read:

• Testing patterns: ./patterns/testing-patterns.md
• Slack patterns: ./patterns/slack-patterns.md
• Environment setup: ./reference/env-vars.md
• AI SDK: ./reference/ai-sdk.md
• Slack setup: ./reference/slack-setup.md
• Vercel deployment: ./reference/vercel-setup.md

Checklist Before Task Completion

Before marking ANY task as complete, verify:

• Code changes have corresponding tests
• pnpm lint passes with no errors
• pnpm typecheck passes with no errors
• pnpm test passes with no failures
• No hardcoded credentials
• Follows existing code patterns
• Chat SDK: Webhook route handles all platforms via bot.webhooks
• Chat SDK: TSConfig includes "jsx": "react-jsx" and "jsxImportSource": "chat" if using JSX components
• Bolt: Events endpoint handles both JSON and form-urlencoded
• Verified AI SDK: using @ai-sdk/gateway (not @ai-sdk/openai) unless user explicitly chose direct provider