Skip to content

internet-court/internet-court-skill / vendored/openserv/openserv-multi-agent-workflows

openserv-multi-agent-workflows

Verified

Multi-agent workflow examples to work together on the OpenServ Platform. Covers agent discovery, multi-agent workspaces, task dependencies, and workflow orchestration using the Platform Client. Read reference.md for the full API reference. Read openserv-agent-sdk and openserv-client for building and running agents.

examples/youtube-to-blog-pipeline.md

View on GitHub
# YouTube to Blog Pipeline

A 3-agent sequential workflow that analyzes a YouTube video, writes a blog post, and generates a header image.

## Pipeline

```mermaid
flowchart LR
    T[Trigger] --> A[Youtube Video Understanding]
    A --> B[Copywriter]
    B --> C[Nano Banana Pro Agent]
```

## Task Dependencies (Sequential)

```mermaid
flowchart TD
    A["Task 1: Youtube Understanding<br/>dependencies: []"] --> B["Task 2: Copywriter<br/>dependencies: [task1Id]"]
    B --> C["Task 3: Nano Banana Pro<br/>dependencies: [task2Id]"]
```

Each task waits for the previous one to complete before starting.

---

## Complete Setup Script

### Project Structure

```
youtube-to-blog-pipeline/
├── src/
│   └── setup.ts
├── .env
├── package.json
└── tsconfig.json
```

### .env

```env
WALLET_PRIVATE_KEY=0x...
```

### Dependencies

```bash
npm init -y && npm pkg set type=module
npm i @openserv-labs/client dotenv
npm i -D @types/node tsx typescript
```

> **Note:** The project must use `"type": "module"` in `package.json`. Add a `"setup": "tsx src/setup.ts"` script for local development.

### src/setup.ts

**Recommended Approach: Using `workflows.sync()`**

```typescript
import 'dotenv/config'
import { PlatformClient, triggers } from '@openserv-labs/client'

async function setup() {
  const client = new PlatformClient()

  if (!process.env.WALLET_PRIVATE_KEY) {
    console.error('Missing WALLET_PRIVATE_KEY in .env')
    process.exit(1)
  }

  console.log('1. Authenticating with wallet...')
  await client.authenticate(process.env.WALLET_PRIVATE_KEY)

  console.log('2. Finding agents from marketplace...')
  const youtubeResult = await client.agents.listMarketplace({ search: 'youtube' })
  const copywriterResult = await client.agents.listMarketplace({ search: 'copywriter' })
  const nanoBananaResult = await client.agents.listMarketplace({ search: 'nano banana' })
  const youtubeUnderstanding = youtubeResult.items[0]
  const copywriter = copywriterResult.items[0]
  const nanoBanana = nanoBananaResult.items[0]

  if (!youtubeUnderstanding || !copywriter || !nanoBanana) {
    console.error('   Could not find required agents')
    const all = await client.agents.listMarketplace({})
    all.items.slice(0, 15).forEach(a => console.log(`   ID: ${a.id} | ${a.name}`))
    process.exit(1)
  }

  console.log(`   Youtube Understanding: ${youtubeUnderstanding.name} (ID: ${youtubeUnderstanding.id})`)
  console.log(`   Copywriter: ${copywriter.name} (ID: ${copywriter.id})`)
  console.log(`   Nano Banana Pro: ${nanoBanana.name} (ID: ${nanoBanana.id})`)

  console.log('3. Creating workflow...')
  const workflow = await client.workflows.create({
    name: 'YouTube to Blog Converter',
    goal: 'Analyze YouTube videos to extract key insights, transform them into comprehensive blog posts, and generate eye-catching header images',
    agentIds: [youtubeUnderstanding.id, copywriter.id, nanoBanana.id],
    triggers: [
      triggers.webhook({
        name: 'webhook',
        waitForCompletion: true,
        timeout: 900,
        input: {
          youtube_url: { type: 'string', title: 'YouTube URL', description: 'YouTube video URL to convert' }
        }
      })
    ],
    tasks: [
      {
        name: 'analyze',
        agentId: youtubeUnderstanding.id,
        description: 'Analyze YouTube video',
        body: 'Analyze the YouTube video at the given URL. Provide a comprehensive summary including: main topics covered, key points and arguments, notable quotes, and overall message.',
        input: '{{trigger.youtube_url}}'
      },
      {
        name: 'write',
        agentId: copywriter.id,
        description: 'Write blog post',
        body: 'Based on the video analysis, write a comprehensive blog post. Structure it with: 1) Engaging headline, 2) Introduction that hooks the reader, 3) Main content sections with subheadings, 4) Key takeaways, 5) Conclusion.'
      },
      {
        name: 'image',
        agentId: nanoBanana.id,
        description: 'Generate blog header image',
        body: 'Generate an eye-catching header image for the blog post. The image should be visually appealing, professional, and relevant to the blog topic.'
      }
    ],
    // ⚠️ CRITICAL: Edges define the data flow
    edges: [
      { from: 'trigger:webhook', to: 'task:analyze' },
      { from: 'task:analyze', to: 'task:write' },
      { from: 'task:write', to: 'task:image' }
    ]
  })
  console.log(`   Workflow ID: ${workflow.id}`)

  console.log('4. Activating workflow...')
  const trigger = workflow.triggers[0]
  await client.triggers.activate({ workflowId: workflow.id, id: trigger.id })
  await workflow.setRunning()

  console.log('\n========================================')
  console.log('YouTube to Blog Pipeline Setup Complete!')
  console.log('========================================')
  console.log(`\nWorkflow ID: ${workflow.id}`)
  console.log(`\nWorkflow: Trigger → Youtube Understanding → Copywriter → Nano Banana Pro`)
  console.log(`\nWebhook URL:`)
  console.log(`  POST https://api.openserv.ai/webhooks/trigger/${trigger.token}`)
  console.log(`\nExample:`)
  console.log(`  curl -X POST https://api.openserv.ai/webhooks/trigger/${trigger.token} \\`)
  console.log(`    -H "Content-Type: application/json" \\`)
  console.log(`    -d '{"youtube_url": "https://www.youtube.com/watch?v=..."}'`)
  console.log('========================================')
}

setup().catch(err => {
  console.error('Setup failed:', err.message)
  process.exit(1)
})
```

---

## How It Works

1. **Trigger fires** with `{ "youtube_url": "https://www.youtube.com/watch?v=..." }`
2. **Youtube Video Understanding Agent** analyzes the video content
3. **Youtube Understanding completes** → Copywriter task becomes ready
4. **Copywriter** writes a blog post
5. **Copywriter completes** → Nano Banana Pro task becomes ready
6. **Nano Banana Pro** generates a header image for the blog
7. **Nano Banana Pro completes** → Blog post and image returned via webhook response

## Workflow Graph

```mermaid
flowchart TD
    T[Trigger] -->|youtube_url| A[Task 1: Youtube Understanding]
    A -->|video analysis| B[Task 2: Copywriter]
    B -->|blog post| C[Task 3: Nano Banana Pro]
    C -->|header image| R[Result]
```

## Usage

```bash
# Run setup (creates workflow, tasks, trigger)
npm run setup

# Trigger the workflow
curl -X POST https://api.openserv.ai/webhooks/trigger/{token} \
  -H "Content-Type: application/json" \
  -d '{"youtube_url": "https://www.youtube.com/watch?v=..."}'
```
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 198kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 198kUpdated Claude CodeCodexVerifiedInstall