Adwarp logo

Adwarp

Create and analyze LinkedIn Ads and Google Ads campaigns, audiences, creatives, and reports through MCP or the adwarp CLI.

Install Adwarp

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "adwarp": {
      "command": "npx",
      "args": [
        "-y",
        "adwarp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

adwarp MCP server connects AI assistants to LinkedIn Ads and Google Ads for campaign creation, audience management, reporting, and ad research. It can resolve targeting, estimate reach, upload matched-audience data, create draft or paused campaigns, and return platform IDs. LinkedIn entities are created as drafts, while Google entities remain paused until enabled in Google Ads. Reach for adwarp when you want conversational campaign workflows or scripted batch operations through any MCP client.

Use cases

  • Create draft LinkedIn campaigns from a brief and audience
  • Build matched audiences from CSV, Salesforce, or HubSpot data
  • Launch paused Google Search campaigns with keywords and responsive ads
  • Compare ad performance and calculate before-and-after lift
  • Inspect competitor ads and summarize their offers

Key features

  • LinkedIn Ads campaign and creative creation
  • Google Search campaign creation
  • Matched-audience uploads and targeting search
  • Performance reporting and lift analysis
  • Competitor ad library inspection
  • Ad change journaling

Tools18

What Adwarp lets your agent do. Expand a tool for its parameters.

Self-reported~76 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 3d ago

Recent health check succeeded.

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 6d ago via OSV.dev · adwarp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars23Stargazers on the source repository.
npm downloads609Package downloads in the last 30 days.
Last commit5d agoMost recent push to the default branch.
Tools exposed18Callable tools this server registers over MCP.

How Adwarp works

What adwarp does

adwarp provides an MCP interface for working with LinkedIn Ads and Google Ads. Its capabilities cover the campaign lifecycle: discovering ad accounts, finding targeting options, estimating audience size, uploading matched-audience data, selecting conversions, creating campaigns and ads, reviewing performance, and recording changes.

LinkedIn workflows include campaign groups, campaigns, text ads, image ads, matched audiences, conversion selection, and performance analysis. Audience data can come from CSV files, Salesforce contacts, or HubSpot lists and searches. The server can also inspect public competitor ads, including copy, formats, and EU targeting metadata.

Google Ads support is focused on search campaigns. A launch can create the budget, campaign, geographic and language targeting, ad groups, keywords, and responsive search ads together. Google campaign creation uses paused entities; LinkedIn campaign entities are drafts. The platform-specific hierarchy differs, so a LinkedIn campaign group and campaign do not map directly to a Google campaign and ad group.

How it works

The primary interface is the @adwarp/mcp package, which runs over stdio for local use. The hosted web package also exposes MCP over HTTP at https://mcp.adwarp.dev/api/mcp. The same underlying engines are available through the adwarp CLI for scripts and batch jobs.

MCP requests can perform operations such as launch_from_brief, performance_summary, or compute_lift. Adwarp resolves the requested account and campaign details, performs the operation, and reports resulting IDs or analysis. Google launch operations validate the request server-side before applying changes and use one atomic mutation, avoiding a partially created campaign when the operation fails.

There is no built-in scheduler. Recurring reports, competitor checks, and audience-status checks must be scheduled by the client or an external system such as cron.

Setup and configuration

Install and run the published npm package or use the hosted MCP endpoint. For local workflows, the package includes the adwarp and adwarp-mcp commands. The CLI and MCP server use the same platform engines.

Google Ads requires a developer token from a manager account and an OAuth desktop client from a Google Cloud project with the Google Ads API enabled. Store Google credentials in ~/.adwarp/google.json, including the client ID, client secret, developer token, login customer ID when accessing an account through a manager, default customer ID, and optional default conversion action names. Complete authentication with adwarp google auth login.

Google keyword research requires Basic developer-token access; campaign creation and reporting work with Explorer access. A Google OAuth consent screen left in Testing produces refresh tokens that expire after seven days.

Tools and capabilities

  • list_ad_accounts lists accessible advertising accounts.
  • list_targeting_facets, search_targeting, and list_facet_entities resolve targeting choices.
  • estimate_audience forecasts reachable audience size before spending.
  • upload_audience_csv cleans and transforms CSV audience data for matching.
  • list_conversions retrieves available conversion actions.
  • create_campaign_group, create_campaign, create_text_ad, and create_image_ad create LinkedIn campaign components.
  • launch_from_brief creates a campaign workflow from a brief.
  • performance_summary and get_performance retrieve performance analysis.
  • inspect_competitor_ads reads public competitor advertising information.
  • log_ad_change, list_ad_changes, and compute_lift support change history and before-and-after analysis.

Limitations and notes

Google support currently covers search campaigns and creation-first workflows. Search network expansion, search partners, and audience expansion are disabled by the platform rules described for Google launches; bidding starts with manual CPC.

New campaigns are intentionally inactive: LinkedIn objects begin as drafts, and Google objects begin paused. LinkedIn campaigns can be enabled through an explicit request or Campaign Manager, while Google campaigns must be enabled in the Google Ads UI. Matched audiences need sufficient members before they can serve; the documented threshold is 300 members.

Adwarp is unofficial and is not affiliated with or endorsed by LinkedIn or Google.

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Adwarp

adwarp is an MCP server that connects AI assistants to LinkedIn Ads and Google Ads. Its main tools manage accounts, targeting, audiences, conversions, campaigns, creatives, performance reports, competitor ad research, change journals, and lift analysis. It runs locally over stdio or can be accessed through a hosted MCP endpoint.

Install the published npm package and run its local MCP command, or connect an MCP client to the hosted endpoint at https://mcp.adwarp.dev/api/mcp. The package also provides the adwarp CLI for scripted and batch workflows.

Yes, adwarp works with Cursor through its standard MCP interface. Use the local stdio server or the hosted streamable HTTP endpoint, then configure Cursor with the selected connection.

Yes, adwarp creates Google Search campaigns with budgets, targeting, ad groups, keywords, and responsive search ads. New Google entities are created paused, and Google Ads credentials require a developer token and OAuth client.

Yes, adwarp can upload cleaned CSV data as a matched audience and build audiences from Salesforce or HubSpot sources. It can also resolve targeting facets and estimate audience reach before campaign creation.

Related MCP servers

Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewAdwarp AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/adwarp?style=directory)](https://allmcps.com/mcp/adwarp)
HTML Embed
<a href="https://allmcps.com/mcp/adwarp"><img src="https://allmcps.com/api/badge/adwarp?style=directory" alt="Adwarp on AllMCPs" /></a>