Blossom Ads
Automatically optimize your marketing creatives with AI. Blossom Ads analyzes performance and scales what works.
https://blossomads.ai/Opens ChatGPT on the web or desktop and asks it to use the WebMCP tools available here.
Connect straight to this server’s public endpoint.
https://mcp.blossomads.ai/mcpWe add this server to your workspace, then open Studio — saved access, one connection to many servers, with a history of what ran.
Last probed Sep 14, 2026 · mcp.blossomads.ai
10tools discovered
Health Check
Liveness probe for the blossomads MCP server. Returns a small structured object so an agent (or a connection test) can confirm the server is reachable and which MCP protocol revision it speaks. Read-only; carries no tenant data.
List Accounts
List the Google Ads accounts this connection is scoped to (read-only).
List Recommendations
List pending replacement recommendations for the scoped account (read-only).
Get Recommendation
Get full detail for one recommendation (read-only, tenant-guarded).
Apply
Apply a recommendation to Google Ads (blossomads executes the mutation). Permitted only for accounts in agent_delegated apply mode (no-double-apply, P1); idempotent (P3). Destructive — surface a confirmation to the human before calling. The account-lookup + defensive Google-client construction + apply_core orchestration (G4) lives in `app.mcp.apply_entry.run_apply_entry` — the SAME function the REST `POST /agent/v1/recommendations/{id}/apply` route calls (P2 plan Task 2.1), so the two surfaces
Generate Recommendations
Ask blossomads to generate new recommendations now (P6 addendum). blossomads decides what is actually due — the per-type pacing budgets, and grace gates are never bypassed by this tool (`force` is hard-wired False; there is no way for a caller to set it). Async: returns `{"status": "queued", ...}` immediately; discover results by re-polling `list_recommendations`/`recs list`. Requires the `pipeline:generate` scope (Operator preset only) AND the account owner having enabled agent-triggered gener
List Asset Scores
List per-asset scores, worst first (read-only, tenant-guarded). Latest per-asset score plus a rolling-window aggregate (Monday 3149170088 — "find assets consistently ranked bottom-5"). `ad_group_id`/ `campaign_id` are Google Ads id strings, tenant-scoped via the account they belong to.
Get Score History
Full per-asset daily score matrix for ONE ad group (read-only, tenant-guarded). A single-pass "find my bottom-N assets over time" surface — one call returns every scored asset's day-by-day quality_score series for the requested ad group. `ad_group_id` is a Google Ads id string.
Get Scores Summary
Group-level score rollup (Monday 3149170088 follow-up): one ad group, one campaign's ad groups, or the whole account's ad groups (default). `level` is one of `"ad_group"`, `"campaign"`, `"account"`; `id` is the Google ad-group/campaign id string, REQUIRED for `level="ad_group"` or `"campaign"`, ignored for `"account"`. Includes materialized label-group virtual ad groups (label-group ads Phase 2), marked `is_label_group`/`label_name` -- addressable at `level="ad_group"` by their own `ad_group_id
Decline
Decline a recommendation; records feedback so future cycles learn.
Get your MCP into directories
A working endpoint is step one. Directory coverage is the coordinated launch across ChatGPT, Claude, Cursor, the MCP Registry, and community indexes.
Directory coverage for brandsAutomatically optimize your marketing creatives with AI. Blossom Ads analyzes performance and scales what works.
Use the MCP endpoint listed on this page in your MCP client configuration. One-click install pills support Claude, Cursor, VS Code, and other hosts. Copy the remote MCP URL if your client needs a manual entry.
Operate Blossom Ads? Verify ownership to take over this directory entry.
This server appears in the MCPBundles directory. Verify you operate it to take over the listing — name, description, logo, contact email, and skill content. We email a 6-digit code to a maintainer address your server publishes in /.well-known/security.txt or /.well-known/mcpbundles.json. Free, takes about a minute.
MCPBundles probed 10 tools on the live server. The tool list on this page reflects what was discovered at the last refresh — connect your client to see the full set available to your session.
No provider sign-in was required during MCPBundles' probe. Your client may still need MCPBundles credentials depending on how you connect.
MCPBundles is an independent platform built on the open Model Context Protocol standard. Not affiliated with Anthropic PBC or Claude.