Adform Media Plan Review

Adform media plan review

Inspect a pre-launch media plan — budget, schedule, goals, targeting — its candidate campaign tree, and Adform's optimisation recommendations. Read-only.

Connection & tooling

Runs on the Adform GraphQL MCP. Use graphql_execute to run queries. Use graphql_introspect on MediaPlan, MediaPlanContent, and MediaPlanRecommendationsFilterInput to discover available fields. Keep calls sequential (~1–2s apart).

---

List media plans

{
  mediaPlans(
    advertiserIds: ["71883"]
    pagination: { offset: 0, limit: 20 }
  ) {
    mediaPlans { id name advertiserId comment schedule { startDate endDate timeZoneId } budget { currency amount } }
    totalCount
  }
}

Get plan detail

{ mediaPlan(id: "b2d428ae-0000-0000-0000-000000000000") { id name advertiserId comment schedule { startDate endDate timeZoneId } budget { currency amount } } }

Use graphql_introspect on MediaPlan to expand goals, KPIs, targeting, schedules, and line items.

Get candidate campaign tree

{ mediaPlanContent(id: "b2d428ae-0000-0000-0000-000000000000") { id } }

Use graphql_introspect on MediaPlanContent to discover the full tree structure — campaigns, orders, and line items the plan would generate.

Get optimisation recommendations

The mediaPlan input requires advertiserId and goals at minimum. Fetch the plan first, then pass its parameters. The filter field is recommendationTypes.

{
  mediaPlanRecommendations(
    mediaPlan: {
      advertiserId: "71883"
      goals: { kpis: [{ performance: { ctr: { rate: 0.005 } } }] }
      schedule: { startDate: "2026-07-01", endDate: "2026-07-31", timeZoneId: "Europe/Berlin" }
    }
    filter: { recommendationTypes: [] }
  ) {
    recommendationGroup
    recommendationType
  }
}

Use graphql_introspect on MediaPlanRecommendationsMediaPlanInput and MediaPlanRecommendationsFilterInput to discover available recommendation types and input fields.

---

Presenting

Summarise the plan: advertiser, flight dates, budget, currency, KPI goals, and key targeting. Present recommendations as a prioritised list explaining what each would change and why a trader might apply or skip it. To estimate what the plan would deliver, use adform-reach-forecast.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: adform
• Source: adform/agentic-skills
• License: MIT
• Homepage: https://site.adform.com

Install and usage instructions live in the source repository linked above.