Meta Ads Manager for Openclaw

A sophisticated AI agent skill designed to manage and optimize Meta (Facebook and Instagram) advertising campaigns through the Meta Marketing API.

aiddun
v1.0.0
Feb 24, 2026
9
4k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install meta-ads-manager

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install meta-ads-manager using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Meta Ads Manager?

The Meta Ads Manager skill transforms your AI agent into a senior Meta Ads strategist capable of managing the full lifecycle of Facebook and Instagram advertising. By integrating with the Metacog MCP server, this skill provides live, authenticated access to ad accounts via secure OAuth, removing the need for manual API key management. It allows users to perform deep-dive performance analysis, adjust budgets, and audit creative quality directly through natural language, making it a cornerstone for those using Openclaw Skills for marketing automation.

Built for efficiency and technical precision, the skill utilizes sandboxed JavaScript to query the Meta Graph API v21.0. It excels at synthesizing raw data into actionable insights, such as calculating ROAS, identifying audience segments with high CPA, and providing specific recommendations for budget reallocation. Whether you are conducting a routine campaign audit or looking for scaling opportunities, this skill provides the tools necessary for data-driven decision-making.

Meta Ads Manager Use Cases

  • Tracking real-time performance metrics like ROAS, CTR, and CPA across multiple ad accounts.
  • Auditing active campaigns to identify and pause underperforming creatives or ad sets.
  • Performing demographic and geographic breakdowns to discover high-value audience segments.
  • Managing campaign statuses and budgets based on automated performance insights.
  • Generating comprehensive weekly or monthly performance reports with trend analysis.

How Meta Ads Manager Works

  1. The skill starts by executing the list_ad_accounts tool to identify all available ad account IDs connected to the user's profile.
  2. Users provide a prompt for analysis or management, which the agent translates into sandboxed JavaScript calls using read_ads or write_ads.
  3. The skill interacts with the Meta Graph API v21.0 using the metaFetch global to retrieve campaign, ad set, and insight data.
  4. Performance data is aggregated and formatted within the sandbox to ensure context efficiency and minimize token usage.
  5. If changes are requested (e.g., pausing a campaign), the agent presents the proposed modifications for user approval before executing metaPost or metaDelete.

Meta Ads Manager Setup

To get started with the Meta Ads Manager in the Openclaw Skills ecosystem, you must have the Metacog MCP server configured. Authentication is handled via a secure OAuth flow.

# Initial command to discover available accounts
# This is typically the first step in any workflow
mcp call meta-ads-manager list_ad_accounts

Meta Ads Manager Data Schema & Taxonomy

The skill organizes data according to the Meta Marketing API structure, focusing on the following entities:

Level Key Metadata & Fields
Campaigns ID, Name, Status, Objective, Daily Budget, Budget Remaining
Ad Sets ID, Name, Targeting, Optimization Goal, Bid Amount
Ads ID, Name, Creative, Quality/Engagement/Conversion Rankings
Insights Spend, Impressions, Clicks, CTR, ROAS, Cost Per Action

Insights can be broken down by dimensions such as age, gender, country, and device_platform using standard API presets or custom date ranges.

Meta Ads Manager Advanced Features

  • Sandboxed JavaScript execution for complex data manipulation and custom metric calculation.
  • Persistence layer via the PERSIST global to maintain state and IDs across multiple tool calls without re-fetching.
  • Parallel fetch execution within the runtime to speed up multi-endpoint data retrieval.
  • Granular creative auditing using quality ranking metrics (Quality, Engagement, and Conversion rate rankings).
  • Write-safety protocols that ensure no campaign mutations occur without explicit manual confirmation.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*