Quick start

1. Get an API key.

   Sign in to the AI4M dashboard and open Settings → API access. Choose Create API key, name it, and pick the data it may read. Copy the key when it is shown: it is displayed only once.

   In an organisation, only administrators can create keys. See Authentication.

2. Check which months are available.

   curl

   curl https://api.ai4mproject.com/v1/periods \
     -H "Authorization: Bearer $AI4M_API_KEY"

   Python

   from ai4m_sdk import AI4MClient
   
   client = AI4MClient(api_key="ai4m_live_...")
   periods = client.get_periods()
   print(periods.last_estimated, periods.last_forecast)

   TypeScript

   import { AI4MClient } from "ai4m-sdk";
   
   const client = new AI4MClient({ apiKey: "ai4m_live_..." });
   const periods = await client.getPeriods();
   console.log(periods.last_estimated, periods.last_forecast);

   Go

   client := ai4m.NewClient("ai4m_live_...")
   ctx := context.Background()
   periods, err := client.Periods(ctx)
   if err != nil {
     log.Fatal(err)
   }
   fmt.Println(periods.LastEstimated, periods.LastForecast)

   The response tells you the latest estimated month and the last forecast month:

   {
     "last_estimated": "2026-07",
     "last_forecast": "2026-11",
     "updated_at": "2026-10-04T00:54:14.000Z",
     "model_version": "ai4m-malaria-1.0"
   }

3. Get risk scores for every state.

   curl

   curl https://api.ai4mproject.com/v1/risk/states \
     -H "Authorization: Bearer $AI4M_API_KEY"

   Python

   report = client.get_state_risk()
   for state in report.states:
       print(state.state, state.score, state.level)

   TypeScript

   const report = await client.getStateRisk();
   for (const state of report.states) {
     console.log(state.state, state.score, state.level);
   }

   Go

   report, err := client.StateRisk(ctx, "")
   if err != nil {
     log.Fatal(err)
   }
   for _, state := range report.States {
     fmt.Println(state.State, state.Score, state.Level)
   }

4. Drill into one state’s LGAs.

   States are identified by a two-letter code, such as KN for Kano. The full list is in States and LGA codes.

   curl

   curl "https://api.ai4mproject.com/v1/risk/lgas?state=KN" \
     -H "Authorization: Bearer $AI4M_API_KEY"

   Python

   kano = client.get_lga_risk("KN")
   print(len(kano.lgas), "LGAs in", kano.state)

   TypeScript

   const kano = await client.getLgaRisk("KN");
   console.log(kano.lgas.length, "LGAs in", kano.state);

   Go

   kano, err := client.LGARisk(ctx, "KN", "")
   if err != nil {
     log.Fatal(err)
   }
   fmt.Println(len(kano.LGAs), "LGAs in", kano.State)

Next steps

• Read Scores and levels so you interpret the numbers correctly.
• Get forecasts for the months ahead.
• Handle errors and limits before you automate anything.