AdMakeAI API for Openclaw

AdMakeAI API helps Openclaw Skills create, edit, analyze, and prepare high-performing ad campaigns across creative formats and Meta advertising workflows.

mesmerlord
v1.1.0
Sep 1, 2026
0
153
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install admakeai

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 admakeai 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 AdMakeAI API?

AdMakeAI API is an Openclaw Skills integration for programmatic advertising workflows, including AI-generated ad images, image editing, ad copy, batch creative testing, UGC video, competitor research, Meta uploads, campaign drafts, and performance analytics. It connects agents to the AdMakeAI REST and MCP APIs through more than 60 tools.

The skill uses brand projects to keep generated creatives aligned with a product's audience, positioning, value proposition, and visual identity. It prioritizes read-only operations by default and requires user confirmation before credit-spending generations or writes to Meta, making Openclaw Skills safer for production marketing automation.

AdMakeAI API Use Cases

  • Generate finished product, promotional, testimonial, lifestyle, comparison, or announcement ads from a concise creative brief.
  • Create prompt-faithful product or lifestyle images when the user wants no automatically added headline, CTA, subline, or logo.
  • Edit existing creatives by changing backgrounds, aspect ratios, styling, or other image attributes.
  • Produce batch ad sets with multiple creative directions for hook and concept testing.
  • Generate ad headlines, hooks, and copy variations based on a completed ad image and brand brief.
  • Create UGC-style video ads and monitor asynchronous video processing.
  • Research competitor advertising through the live Meta Ad Library and save inspiration for future remixes.
  • Browse, search, paginate, favorite, and audit previously generated ads and ad sets.
  • Validate creatives before Meta upload, prepare Meta campaign drafts, and inspect connected Facebook ad accounts.
  • Review account, campaign, ad set, ad, spend, CTR, CPC, conversion, ROAS, and public comment analytics.
  • Check AdMakeAI credit balances before large batches or other credit-consuming workflows.

How AdMakeAI API Works

  1. Authenticate with the user's ADMAKEAI_API_KEY through the REST API or expose the same procedures through MCP.
  2. Resolve the correct brand project with adResearch.listProjects before generating creative. If multiple projects match, identify the selected project; if the request is ambiguous, ask the user rather than guessing.
  3. Read the project brief with adResearch.getProject when audience, positioning, product, or brand details affect the request.
  4. Route the user's intent to the appropriate tool, such as adGeneration.create for finished ads, adGeneration.generateImage for exact prompts, adGeneration.edit for image-to-image changes, or adSet.generateBatch for variations.
  5. Upload product images when needed with upload.uploadImageBase64, then pass the returned image ID as inputImageId or use project asset URLs as references.
  6. Estimate credit usage and call user.getCredits before batches larger than 10 generations; request confirmation before any credit-spending or Meta write action.
  7. Submit asynchronous generations and return the operation identifier, status, and expected workflow to the user.
  8. Poll generation or research status with the relevant status endpoint until completion, typically every 3–5 seconds for image generation.
  9. Generate copy, favorite winning concepts, run Meta preflight checks, or assemble campaign drafts from completed assets.
  10. For Meta uploads or draft pushes, read the intended plan back to the user and obtain explicit confirmation before calling the write procedure.
  11. Present analytics and results using the appropriate account, campaign, timeline, ad, or comment reporting tool.

AdMakeAI API Setup

  1. Create an AdMakeAI account and configure a brand project in the dashboard. Projects cannot be created through this API.
  2. Generate an API key at https://admakeai.com/dashboard/integrations/api.
  3. Store the key as an environment variable and never print it, echo it, or write it into a user-visible file.
export ADMAKEAI_API_KEY="amai_live_..."
  1. Confirm that curl is available and that the agent has network access.
curl --version
curl -H "x-api-key: $ADMAKEAI_API_KEY" \
  https://admakeai.com/api/v1/user.me
  1. Use the REST base URL https://admakeai.com/api/v1 or the MCP endpoint https://admakeai.com/api/mcp. Review typed inputs and examples at https://admakeai.com/api/docs.
  2. Connect a Facebook account in the AdMakeAI dashboard before using Facebook account, upload, campaign draft, or Meta reporting tools:
open https://admakeai.com/dashboard/integrations/facebook
  1. Configure the Openclaw Skills runtime with the skill metadata requirements: ADMAKEAI_API_KEY and the curl binary. The declared tools are Bash, Read, Write, and WebFetch.

AdMakeAI API Data Schema & Taxonomy

API and transport structure

Area Structure
REST base URL https://admakeai.com/api/v1
MCP URL https://admakeai.com/api/mcp
Procedure naming Canonical tRPC dotted paths such as adGeneration.create
MCP naming Dots become double underscores, such as adGeneration__create
REST response { ok, tool, data } wrapper
MCP response Result is returned inside content[0].text

Core entities

  • Project: Brand or product record containing product summary, audience, value proposition, logo, assets, competitors, and research state. Generation tools use projectId.
  • AdGeneration: Image creative record with an ID, prompt, status, credit cost, image URL, quality, aspect ratio, and optional source or reference information.
  • AdSet: Shared project, brief, ad type, quality, aspect ratio, and a collection of generated variations.
  • AdCopy: Copy generation output containing up to five hook, headline, or messaging variations tied to a completed image and project.
  • VideoGeneration: Asynchronous UGC or video record with processing status, quality, project scope, and resulting media.
  • Inspiration and scraped ads: Competitor research records that can be saved and referenced for creative remixing.
  • FacebookConnection: Connected Meta identity containing ad accounts and Pages.
  • CampaignDraft and upload records: Reusable creative assets, draft campaigns, ad sets, ads, Meta IDs, and upload states.
  • Analytics: Account, campaign, ad set, and ad performance data including spend, impressions, clicks, CTR, CPC, conversions, ROAS, timelines, and comments.

Common metadata

  • aspectRatio: 1:1, 4:5, 9:16, or 16:9; useful for feed, Reels, Stories, YouTube, and desktop placements.
  • quality: low, high, or ultra, with higher levels consuming more credits.
  • adType: product_showcase, promotional, testimonial, lifestyle, comparison, or announcement.
  • status: Asynchronous records commonly use PENDING and COMPLETED states.
  • Pagination uses { page, perPage }, { limit, cursor }, or { limit, offset } depending on the procedure.

Credit taxonomy

Operation Credit behavior
Image generation or editing 2 credits low, 8 high, 20 ultra
Batch generation Number of variations multiplied by per-generation cost
UGC video 5 low, 15 medium, 50 high per video
Meta upload 3 credits per upload
Reads, analytics, project updates, uploads, copy, and research Free unless a procedure specifies otherwise

Failed generations are refunded automatically. The API intentionally exposes no delete operations, and Meta pause or resume procedures are disabled and fail closed.

AdMakeAI API Advanced Features

  • Dual renderer support: use adGeneration.create for brand-aware finished ads or adGeneration.generateImage for an exact, fully controlled image prompt.
  • Brand-context expansion: project briefs automatically inform finished ad prompts, including headline, subline, CTA, and available logo placement.
  • Reference-driven generation using uploaded product photos, project assets, reference image URLs, source ads, and saved competitor inspiration.
  • Batch creative testing through shared ad set briefs, per-variation directions, and free AI-generated suggestions from adSet.suggestDirections.
  • Asynchronous job management for image generation, UGC video, and competitor scraping with dedicated polling endpoints.
  • Competitor intelligence through company search, Meta Ad Library retrieval, competitor scraping, longest-running ad sorting, and inspiration libraries.
  • REST and MCP interoperability, allowing the same allowlisted procedures to be called through HTTP or MCP tool names.
  • Cursor, offset, and page-based pagination with controlled fetching to avoid unnecessary API traffic.
  • Meta workflow safeguards, including free preflight validation, explicit confirmation for uploads and campaign pushes, and read-only defaults.
  • Multi-stage analytics covering account overviews, period comparisons, campaign trees, timelines, ad-level metrics, and public comments.
  • Credit-aware automation that checks balances before large batches, explains quality-based costs, and warns about daily generation limits.
  • Operational safety constraints that prevent project guessing, prohibit exposed API keys, avoid unsupported deletion, and direct status changes to Meta Ads Manager.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*