azureAgentsStorageAdapter
Stores Assistants-surface state on Azure AI Foundry. Azure owns assistants, threads, and runs; Prisma covers durability and run-step caching.
Signature
azureAgentsStorageAdapter({
azureAiProject: AIProjectClient,
prisma: PrismaClient,
})
Install
npm install supercompat openai @azure/ai-projects @azure/identity @prisma/client
npm install -D prisma
Prisma tables
Because this adapter relies on Prisma, you need the same tables
prismaStorageAdapter uses for the Assistants API plus one extra model that persists function-tool outputs across Azure run turns:Add the Assistants API schema to your
schema.prisma.Add the
AzureAgentsFunctionOutput model in the same file.Apply the changes:
npx prisma db push
npx prisma generate
Example
import { AIProjectClient } from '@azure/ai-projects'
import { DefaultAzureCredential } from '@azure/identity'
import { PrismaClient } from '@prisma/client'
import {
supercompat,
azureAiProjectClientAdapter,
azureAgentsRunAdapter,
azureAgentsStorageAdapter,
} from 'supercompat/openai'
const prisma = new PrismaClient()
const azureAiProject = new AIProjectClient(
process.env.AZURE_AI_PROJECT_ENDPOINT!,
new DefaultAzureCredential(),
)
const client = supercompat({
clientAdapter: azureAiProjectClientAdapter({ azureAiProject }),
storageAdapter: azureAgentsStorageAdapter({ azureAiProject, prisma }),
runAdapter: azureAgentsRunAdapter({ azureAiProject }),
})
const assistant = await client.beta.assistants.create({
model: 'my-agent-deployment',
instructions: 'You are a concise assistant.',
})
const thread = await client.beta.threads.create()
await client.beta.threads.messages.create(thread.id, { role: 'user', content: 'Hello.' })
const run = await client.beta.threads.runs.createAndPoll(thread.id, { assistant_id: assistant.id })