The MCP tool directory.

[write] Star or unstar one of the user's ad generations. Free — use it to mark the winner of a batch.

Allow test end users on one of your application's production realm. Destructive: needs a credential holding the destructive class (a signed-in session, or a token minted with `destructive`); the call creates a pending action, and the admission stands only after a person approves it in the browser. Approving it means any credential of your account that reaches the application can sign a test end user in on production with no mailbox; their sign-ins and verifications meter as backend actions. The admission lasts until `withdraw_test_end_users` and shows as `test_end_users_admitted_at` in `read_realm`, `configure_realm`, and `export_account`. The development realm needs no admission. Refused `test_end_users_not_admitted` under the workforce audience and `test_end_users_disabled` while the platform's `TEST_END_USERS` setting is off.

[write] Add competitors to a project by name (optionally with website URL) and kick off ad scraping for them.

Get one scraped competitor ad in full: creative URLs, ad text, headline, run dates, and platform placements.

Get the ads a Facebook Page is currently running, from the Meta Ad Library, by pageId (from adResearch.searchCompanies). Works for any advertiser, not just the user's competitors.

Preview the ads already scraped for a single competitor.

Get one project with its full brand brief, competitors, research jobs, and scraped-ad counts. Read this before generating so the creative matches the brand.

List the competitor ads saved as inspiration for a project, with their inspiration IDs.

Browse the competitor ads scraped into a project, with sorting (relevance, longest_running, newest) and media-type/keyword filters. This is the raw material for remixes.

List the user's projects (brands). ALWAYS call this before generating anything: every generation needs a projectId, and the project's brand brief is what the render is built from. Pick the project the user's request is about; if more than one could fit, ask which. Projects cannot be created through this API — if the user has none, tell them to set one up at https://admakeai.com/dashboard.

Poll a project's research status: pipeline stage, running scrape jobs, competitor and saved-ad counts.

[write] Save a scraped competitor ad to a project as inspiration. The returned inspiration ID can be passed to adGeneration.create as referenceInspirationId to remix it.

[write] Re-scrape one competitor's live Facebook ads into the project. Returns a research job — poll adResearch.pollStatus.

Resolve a brand name to its Facebook advertiser Page (id, name, category, follower counts). Step 1 of looking up any brand's live ads.

[write] Create an ad set: a project + shared brief + ad type + quality + aspect ratio that many variations are generated from. Free — the credits are charged by adSet.generateBatch.

[write] Generate a batch of ad variations from one ad set + per-variation instructions. Enqueues N background jobs and returns their generation IDs.

Get an ad set by ID, including its general instructions.

List all ad generations belonging to an ad set.

Paginated list of the user's ad sets.

[write] Ask the model for distinct creative directions (angles/hooks) for an ad set, to feed back in as per-variation instructions.

[write] Update an ad set's brief, ad type, quality, aspect ratio, or reference images before generating another batch.

Conversational AI that guides analysis and interprets results.

Zero-context onboarding playbook for agents using this MCP server. Returns the recommended call sequence, common guardrails, and recovery hints.

Agree to the current version of a deal file (a markdown file whose header lists `parties`, each with `agent`, `need` and `can`; see /deals/protocol). Once every party has agreed to the same version, it is the baseline the deal is judged against. Editing the file later doesn't change what was agreed.

Search software on AlternativeTo. Compiled from alternativeto.net's own requests and shared by an Unbrowse user. Inputs: query. Returns the page's title, readable text and links. Read-only on alternativeto.net.

Browse apps by category. Compiled from alternativeto.net's own requests and shared by an Unbrowse user. No inputs. Returns platformsFilters, featureFilters, licenseFilters, originFilters. Read-only on alternativeto.net.

Search alternativeto. Compiled from alternativeto.net's own requests and shared by an Unbrowse user. Inputs: query. Returns results. Read-only on alternativeto.net.

Read a alternativeto.net software page. Compiled from alternativeto.net's own requests and shared by an Unbrowse user. Inputs: software. Returns the page's title, readable text and links. Read-only on alternativeto.net.

Search apps and software. Compiled from alternativeto.net's own requests and shared by an Unbrowse user. Inputs: query. Returns the page's title, readable text and links. Read-only on alternativeto.net.

Analyze the citation network for a specific paper