Give your agent marketing superpowers

How people use it

What a founder types to their coding agent, and what happens underneath. Every example maps to one of the fifty-four tools above.

• "What should I post tomorrow for my meditation app?"

  Layers find_growth_opportunities returns the standing ranked queue, each test carrying the post it came from and how that post beat its own account's normal.

• "Which of our competitors' posts broke out this week?"

  Layers manage_growth_landscape keeps the tracked set and layers://landscape returns it ranked, with the source evidence and when it was observed.

• "Plan a three-hook test for our onboarding video and send the winner to TikTok drafts."

  Layers plan_content_experiment writes the plan with one controlled variable, render_content makes the variants, and deliver_content_experiment puts the winner in the draft inbox.

• "Track the video I posted from my own account and tell me if it beat our normal."

  Layers track_growth_experiment registers the post with its hypothesis, and review_growth_results scores it against the account's own top quartile.

• "Review last week's experiment and tell me what to do more of."

  Layers review_growth_results reads the evidence, records the verdict with its uncertainty, and names the next controlled test to run.

• "Stop everything on this product now."

  Layers halt stops queued and running work in the scope you named and returns a receipt for every target it touched.

• "Make me a face for the app and put her in a reaction video."

  Layers generate_influencer creates the persona for free and hands back a slate page of every face in the project, then render_content casts the completed one in a reaction_demo at two hundred credits.

• "That price looks wrong to me. Tell the Layers team."

  Layers send_feedback goes to the founders, who read every message. It is free and it works even when the balance is empty.

• "Set Layers up for this repo and connect our TikTok account."

  Layers onboard_product derives the brief from the repository, the website and the store listing, then returns the connection page. It sends the brief fields and a digest. No file, path or diff leaves your machine.

• "Read the repo, write our product brief, and tell me who we are actually competing with for attention."

  Layers onboard_product writes the reviewed brief and starts the first landscape build and the project's first assets, all included at no charge.

• "Add these four developer accounts to the map and drop the ones that are not in our category."

  Layers manage_growth_landscape pins and removes accounts and rescopes what counts as direct, adjacent or inspirational.

• "Turn our changelog post into a short-form test, and tell me the credits before you spend any."

  Layers plan_content_experiment and render_content both accept quoteOnly, which returns the exact maximum credits and the price version before any work runs.

• "Draft this week's post about the new integration and hold it for me."

  Layers deliver_content_experiment prepares the artifact in draft mode, which ends in the platform draft inbox and posts nothing publicly. You make the final public action there.

• "Did the pricing-page hook beat the feature hook?"

  Layers review_growth_results reads both experiments against the win gate and writes the verdict, then layers://experiments holds the append-only ledger behind it.

• "Cancel the trial renewal before Friday."

  Layers cancel_trial_auto_renewal previews the trial end, the effective date, what you keep and what you lose, and returns the link.

• "Promote the demo clip that won last month."

  Layers plan_paid_campaign refuses every platform today: no paid lane is certified. The Paid Growth lane ledger records when a TikTok, Apple, or Google lane is live-certified, and the planner takes that platform again on the same day.

• "We have one account and no time. Run five more for us."

  Layers plan_managed_distribution quotes five Layers-operated TikTok accounts at $150 each a month against your cash wallet, and provision_approved_managed_accounts orders exactly that fleet.

• "What should we post for the summer drop next week?"

  Layers find_growth_opportunities ranks the current tests with the evidence behind each one, and plan_content_experiment turns the one you pick into a plan.

• "Find the creators whose posts about products like ours are breaking out, and start tracking them."

  Layers manage_growth_landscape pins them into the tracked set, and every later opportunity cites the post it came from.

• "Turn the product shots into a slideshow and put it in TikTok drafts."

  Layers render_content builds it with the slideshow_builder recipe at eighty five credits per usable variant, composing the project's own screens by their ids from layers://brand-assets, and deliver_content_experiment prepares it with its disclosures and caption.

• "This creator's TikTok is doing our product's job better than we are. Make our version."

  Layers render_content with the adapt_source recipe remixes that public post: two hundred credits for a ten second video, four hundred for twenty seconds, eighty five when the source is a photo post and the remix is a deck.

• "Launch a TikTok test for the summer drop and end it on the 14th."

  Layers The same answer: plan_paid_campaign refuses the platform, so create_approved_campaign_draft and launch_approved_campaign have no plan to act on. Nothing drafts and nothing spends until the lane ledger records a live-certified row.

• "Cut the budget on the campaign that is not converting."

  Layers apply_approved_campaign_change previews the bounded change, then applies exactly what the preview showed.

• "Register the video we posted from the brand account and score it against our normal."

  Layers track_growth_experiment binds the post to a hypothesis, and review_growth_results scores it once there is enough evidence.

• "Pause every campaign and every queued post right now."

  Layers halt stops future work and pauses active spend inside the scope you named, and returns a receipt per target.

Manual configuration

The paste at the top has your agent do all of this. Here are the same steps for a shell you drive yourself, and the wiring for a client you configure by hand.

npm install -g @layers/cli
layers setup

{
  "mcpServers": {
    "layers": { "command": "layers", "args": ["mcp"] }
  }
}

POST https://mcp.layers.com/mcp
Authorization: Bearer <layers session>
Content-Type: application/json
Accept: application/json, text/event-stream