# Send Ad Context Protocol (AdCP) Advertising to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "adcp-advertising",
    "name": "Ad Context Protocol (AdCP) Advertising",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/edyyy62/adcp-advertising",
    "canonicalUrl": "https://clawhub.ai/edyyy62/adcp-advertising",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/adcp-advertising",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=adcp-advertising",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "CREATIVE.md",
      "EXAMPLES.md",
      "REFERENCE.md",
      "TARGETING.md",
      "README.md",
      "SKILL.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-23T16:43:11.935Z",
      "expiresAt": "2026-04-30T16:43:11.935Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
        "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/adcp-advertising"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/adcp-advertising",
    "downloadUrl": "https://openagent3.xyz/downloads/adcp-advertising",
    "agentUrl": "https://openagent3.xyz/skills/adcp-advertising/agent",
    "manifestUrl": "https://openagent3.xyz/skills/adcp-advertising/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/adcp-advertising/agent.md"
  }
}
```
## Documentation

### Overview

Automate your advertising campaigns with AI. This skill enables OpenClaw agents to discover ad inventory, launch campaigns, manage creatives, and optimize performance across display, video, CTV, audio, and more - all through natural language commands.

No dashboards. No forms. No ad platform expertise required.

### What You Can Do

🎯 Launch campaigns in minutes - "Create a $10k display campaign targeting tech professionals in California"
🔍 Discover ad inventory instantly - "Find premium video placements for luxury brands"
🎨 Upload ads with ease - "Upload these banner images as creatives"
📊 Track ROI in real-time - "Show me campaign performance and CTR by creative"
🎛️ Auto-optimize spend - "Reallocate budget to top-performing packages"
🌐 Target precisely - Demographics, behaviors, interests, locations, devices, times

### Perfect For

Marketing teams running Facebook ads, Google ads, and multi-channel campaigns
Media buyers managing programmatic ad spend across publishers
Agencies automating client campaign management and reporting
E-commerce brands launching product ads and retargeting campaigns
Startups running lean marketing with AI-powered automation

### Why Choose This Skill?

Skip the learning curve - No need to master complex ad platforms
Save time - Launch in 5 minutes vs. hours of manual setup
Spend smarter - AI automatically optimizes budgets to top performers
Scale faster - Manage unlimited campaigns through simple commands
Test risk-free - Public test agent included, no setup required

Official AdCP Repository: https://github.com/adcontextprotocol/adcp
Official AdCP Documentation: https://docs.adcontextprotocol.org
Complete Documentation Index: https://docs.adcontextprotocol.org/llms.txt

### When to Use This Skill

Trigger this skill when users ask about:

Campaign Management

"Create a display ad campaign"
"Launch Facebook ads for my product"
"Set up a $5000 video advertising campaign"
"Pause my underperforming campaigns"

Ad Discovery & Media Buying

"Find advertising inventory for luxury brands"
"Show me CTV ad placements in major cities"
"What display ad options are available?"
"Buy media for a tech startup"

Creative Management

"Upload these banner images"
"Which creative is performing best?"
"Add video ads to my campaign"
"Manage my ad library"

Performance & Optimization

"How is my campaign performing?"
"Show me ROI by channel"
"Optimize my ad spend"
"Reallocate budget to top performers"
"Track impressions and click-through rates"

Targeting & Audiences

"Target professionals in California"
"Set up demographic targeting"
"Create a retargeting campaign"
"Target by device type and time of day"

### Launch Your First Campaign (5 Minutes)

No setup required. Use the included test agent to try everything:

Step 1: Discover what's available

"Show me advertising capabilities"

Browse available channels, publishers, and formats.

Step 2: Find ad inventory

"Find display ads for a tech startup, budget $5000"

AI searches and shows matching products with pricing.

Step 3: Launch campaign

"Create campaign with Product prod_123, $5000 budget, targeting California tech professionals"

Campaign goes live instantly.

Step 4: Upload your ads

"Upload these banner images as creatives"

Drop files, get instant creative IDs.

Step 5: Monitor performance

"Show campaign metrics and ROI"

Real-time impressions, clicks, CTR, spend.

### Real-World Usage Examples

Quick campaign launch:

User: "I need to run display ads for my SaaS product"
Agent: [Discovers products] "Found 5 display packages. Want details?"
User: "Create campaign with Product 1, $10k budget, target CTOs"
Agent: [Creates campaign] "Campaign live! ID: mb_abc123"

Performance optimization:

User: "How are my video ads performing?"
Agent: [Shows metrics] "Package A: 2.3% CTR, Package B: 0.8% CTR"
User: "Move $5k from B to A"
Agent: [Reallocates] "Budget updated. Package A now $15k"

Multi-channel campaign:

User: "Launch omnichannel campaign: display in CA, video in NYC, $50k total"
Agent: [Creates packages] "3 packages created across display and video"

### Natural Language Understanding

Speak naturally. The skill understands:

Budgets: "$5000", "five thousand dollars", "5k budget"
Locations: "California", "major US cities", "New York and LA"
Audiences: "tech professionals", "age 25-45", "high income"
Goals: "brand awareness", "drive conversions", "increase sales"

### Progressive Workflow

1. Discovery Phase

"Find video advertising for luxury brands"

↓ Agent searches inventory
↓ Shows matched products with pricing
↓ Explains targeting and formats

2. Campaign Creation

"Create campaign with Product 1, $25k, target professionals"

↓ Agent creates media buy
↓ Sets up targeting overlay
↓ Returns campaign ID and status

3. Creative Management

"Upload my banner ads"

↓ Agent syncs creatives
↓ Assigns to campaign
↓ Returns creative IDs

4. Monitoring & Optimization

"Show performance"

↓ Agent fetches delivery data
↓ Shows metrics by package/creative
↓ Suggests optimizations

### Create Campaign

const campaign = await testAgent.createMediaBuy({
  buyer_ref: 'campaign-2026-q1',
  brand_manifest: { url: 'https://acme.com' },
  packages: [{ product_id: 'premium_display', budget: 10000 }]
});

### Upload Creatives

await testAgent.syncCreatives({
  creatives: [{ 
    buyer_ref: 'banner-300x250',
    url: 'https://cdn.acme.com/banner.jpg'
  }]
});

### Monitor Performance

const delivery = await testAgent.getMediaBuyDelivery({
  media_buy_id: 'mb_abc123'
});
console.log(\`CTR: ${delivery.totals.ctr}%, Spend: $${delivery.totals.spend}\`);

See REFERENCE.md for complete API docs and EXAMPLES.md for detailed workflows.

### The 8 Media Buy Tasks

AdCP provides 8 standardized tasks for the complete advertising lifecycle. Learn more in the Media Buy Protocol documentation.

get_adcp_capabilities - Discover agent features and portfolio (~1s)
get_products - Find inventory using natural language (~60s)
list_creative_formats - View creative specifications (~1s)
create_media_buy - Launch campaigns (minutes-days, may require approval)
update_media_buy - Modify campaigns (minutes-days)
sync_creatives - Upload creative assets (minutes-days)
list_creatives - Query creative library (~1s)
get_media_buy_delivery - Track performance (~60s)

Complete task reference: https://docs.adcontextprotocol.org/docs/media-buy/task-reference/

### Brand Manifest

Brand context can be provided two ways:

URL reference (recommended - agent fetches brand info):

{
  "brand_manifest": {
    "url": "https://brand.com"
  }
}

Inline manifest (full brand details):

{
  "brand_manifest": {
    "name": "Brand Name",
    "url": "https://brand.com",
    "tagline": "Brand tagline",
    "colors": { "primary": "#FF0000" },
    "logo": { "url": "https://cdn.brand.com/logo.png" }
  }
}

### Pricing Models

Products support various pricing models:

CPM (Cost Per Mille/Thousand) - Fixed price per 1000 impressions
CPM-Auction - Bid-based pricing for impressions
CPCV (Cost Per Completed View) - Video completions
Flat-Fee - Fixed campaign cost
CPP (Cost Per Point) - Percentage of audience reached

For auction pricing, include bid_price in your package.

### Asynchronous Operations

AdCP is not a real-time protocol. Operations may take:

~1 second - Simple lookups (formats, creative lists)
~60 seconds - AI/inference operations (product discovery)
Minutes to days - Operations requiring human approval (campaign creation)

Always check the status field in responses:

completed - Operation finished successfully
pending - Awaiting approval or processing
failed - Operation failed (check error details)

### Targeting Capabilities

Apply targeting overlays to campaigns:

{
  targeting_overlay: {
    geo: {
      included: ['US-CA', 'US-NY'],  // DMA codes or regions
      excluded: ['US-TX']
    },
    demographics: {
      age_ranges: [{ min: 25, max: 44 }],
      genders: ['M', 'F']
    },
    behavioral: {
      interests: ['technology', 'gaming'],
      purchase_intent: ['consumer_electronics']
    },
    contextual: {
      keywords: ['innovation', 'design'],
      categories: ['IAB19'] // Technology & Computing
    }
  }
}

### Workflow 1: Campaign Discovery to Launch

// 1. Discover capabilities
const caps = await agent.getAdcpCapabilities({});

// 2. Find products
const products = await agent.getProducts({
  brief: 'Q1 2026 brand awareness campaign for tech startup',
  brand_manifest: { url: 'https://startup.com' },
  filters: { channels: ['display', 'video'] }
});

// 3. Check creative formats
const formats = await agent.listCreativeFormats({
  format_types: ['display', 'video']
});

// 4. Create campaign
const campaign = await agent.createMediaBuy({
  buyer_ref: 'q1-2026-awareness',
  brand_manifest: { url: 'https://startup.com' },
  packages: [
    {
      buyer_ref: 'pkg-001',
      product_id: products.products[0].product_id,
      pricing_option_id: 'cpm-standard',
      budget: 15000
    }
  ],
  start_time: { type: 'asap' },
  end_time: '2026-03-31T23:59:59Z'
});

// 5. Upload creatives
await agent.syncCreatives({
  creatives: [...], // Your creative assets
  assignments: {
    'creative_001': ['pkg-001']
  }
});

// 6. Monitor performance
const delivery = await agent.getMediaBuyDelivery({
  media_buy_id: campaign.media_buy_id
});

### Workflow 2: Update Running Campaign

// Pause, adjust budget, and resume campaign
await agent.updateMediaBuy({
  media_buy_id: 'mb_abc123',
  updates: {
    status: 'paused',
    budget_change: 5000, // Add $5000
    end_time: '2026-04-30T23:59:59Z'
  }
});

// Resume after adjustments
await agent.updateMediaBuy({
  media_buy_id: 'mb_abc123',
  updates: { status: 'active' }
});

More workflow examples: See EXAMPLES.md for complete campaign scenarios including creative management, multi-channel campaigns, and optimization workflows.

### Test Agent

For development and testing, use the public test agent:

Agent URL: https://test-agent.adcontextprotocol.org/mcp
Auth Token: 1v8tAhASaUYYp4odoQ1PnMpdqNaMiTrCRqYo9OJp6IQ

import { testAgent } from '@adcp/client/testing';

// No authentication needed for test agent
const result = await testAgent.getProducts({
  brief: 'Test campaign',
  brand_manifest: { url: 'https://example.com' }
});

Interactive testing available at: testing.adcontextprotocol.org

### Error Handling

Common error patterns:

400 Bad Request - Invalid parameters:

{
  "error": {
    "code": "VALIDATION_ERROR",
    "message": "budget must be greater than 0",
    "field": "packages[0].budget"
  }
}

401 Unauthorized - Missing or invalid auth:

{
  "error": {
    "code": "UNAUTHORIZED",
    "message": "Invalid authentication token"
  }
}

404 Not Found - Invalid ID reference:

{
  "error": {
    "code": "NOT_FOUND",
    "message": "Product not found",
    "resource": "product_id: premium_video_30s"
  }
}

Always check for errors before processing responses:

if (result.error) {
  console.error(\`Error: ${result.error.message}\`);
  return;
}

### 1. Always Start with Capabilities

Call get_adcp_capabilities first to understand what the agent supports before making other requests.

### 2. Use Clear Buyer References

Use descriptive buyer_ref values for tracking:

Good: 'campaign-2026-q1-tech-launch'
Avoid: 'c1', 'test', 'abc'

### 3. Handle Async Operations

Check status field and implement polling for pending operations:

let status = 'pending';
while (status === 'pending') {
  await sleep(5000); // Wait 5 seconds
  const update = await agent.getMediaBuyDelivery({
    media_buy_id: campaign.media_buy_id
  });
  status = update.status;
}

### 4. Write Detailed Briefs

Better briefs lead to better product matches:

Good: 'Premium video inventory for luxury automotive brand targeting high-income professionals aged 35-54 in major metros. Focus on brand awareness with completion rates above 70%.'
Avoid: 'video ads', 'need advertising'

### 5. Validate Creative Formats

Always check list_creative_formats to ensure your creatives meet requirements before uploading.

### 6. Monitor Budget Pacing

Regularly check delivery metrics to ensure campaigns are pacing properly:

const delivery = await agent.getMediaBuyDelivery({
  media_buy_id: campaign.media_buy_id
});

const pacing = delivery.delivery.spend / delivery.delivery.budget;
console.log(\`Budget pacing: ${(pacing * 100).toFixed(1)}%\`);

### Official AdCP Documentation

Main Documentation: https://docs.adcontextprotocol.org
Complete Index: https://docs.adcontextprotocol.org/llms.txt
Media Buy Protocol: https://docs.adcontextprotocol.org/docs/media-buy/
Quick Reference: https://docs.adcontextprotocol.org/docs/media-buy/quick-reference
Task Reference: https://docs.adcontextprotocol.org/docs/media-buy/task-reference/
Quickstart Guide: https://docs.adcontextprotocol.org/docs/quickstart

### This Skill's Documentation

REFERENCE.md - Complete API reference and schemas
EXAMPLES.md - Real-world campaign examples
PROTOCOLS.md - MCP vs A2A protocol details
TARGETING.md - Advanced targeting strategies
CREATIVE.md - Creative asset management guide

### Key Reminders

AdCP is asynchronous - Operations may take minutes to days
Human approval may be required - Check for pending status
Start with capabilities - Always call get_adcp_capabilities first
Brand context matters - Provide detailed brand manifests for better results
Targeting is additive - Product targeting + your overlay = final targeting
Creative formats are strict - Always validate against format specifications
Monitor performance - Regular delivery checks ensure campaign success

### Support

For help with AdCP:

Official Repository: https://github.com/adcontextprotocol/adcp
Documentation: https://docs.adcontextprotocol.org
Interactive Testing: https://testing.adcontextprotocol.org
Complete API Docs: https://docs.adcontextprotocol.org/llms.txt
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: edyyy62
- Version: 1.0.1
## Source health
- Status: healthy
- Source download looks usable.
- Yavira can redirect you to the upstream package for this source.
- Health scope: source
- Reason: direct_download_ok
- Checked at: 2026-04-23T16:43:11.935Z
- Expires at: 2026-04-30T16:43:11.935Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/adcp-advertising)
- [Send to Agent page](https://openagent3.xyz/skills/adcp-advertising/agent)
- [JSON manifest](https://openagent3.xyz/skills/adcp-advertising/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/adcp-advertising/agent.md)
- [Download page](https://openagent3.xyz/downloads/adcp-advertising)