Deep Research Agent on Supabase Edge Functions

A Research Agent powered by Resonate and OpenAI, running on Supabase Edge Functions. The Research Agent is a distributed, recursive agent that breaks a research topic into subtopics, researches each subtopic recursively, and synthesizes the results.

How It Works

This example demonstrates how complex, distributed agentic applications can be implemented with simple code in Resonate's Distributed Async Await: The research agent is a recursive generator function that breaks down topics into subtopics and invokes itself for each subtopic:

function* research(ctx, topic, depth) {
  const messages = [
    { role: "system", content: "Break topics into subtopics..." },
    { role: "user", content: `Research ${topic}` }
  ];

  while (true) {
    // Ask the LLM about the topic
    const response = yield* ctx.run(prompt, messages, ...);
    messages.push(response);

    // If LLM wants to research subtopics...
    if (response.tool_calls) {
      const handles = [];

      // Spawn parallel research for each subtopic
      for (const tool_call of response.tool_calls) {
        const subtopic = ...;
        const handle = yield* ctx.beginRpc(research, subtopic, depth - 1);
        handles.push([tool_call, handle]);
      }

      // Wait for all subtopic results
      for (const [tool_call, handle] of handles) {
        const result = yield* handle;
        messages.push({ role: "tool", ..., content: result });
      }
    } else {
      // LLM provided final summary
      return response.content;
    }
  }
}

The following video visualizes how this recursive pattern creates a dynamic call graph, spawning parallel research branches that fan out as topics are decomposed, then fan back in as results are synthesized:

call-graph.mov

Key concepts:

• Concurrent Execution: Multiple subtopics are researched concurrently via ctx.beginRpc
• Coordination: Handles are collected first, then awaited together (fork/join, fan-out/fan-in)
• Depth control: Recursion stops when depth reaches 0

---

Running the Example

You can run the Deep Research Agent locally on your machine with Supabase CLI or you can deploy the agent to Supabase Platform.

1. Running Locally

1.1. Prerequisites

Install the Resonate Server & CLI with Homebrew or download the latest release from Github.

brew install resonatehq/tap/resonate

Install the Supabase CLI

To run this project you will also need to add an OpenAI API Key to the supabase/functions/.env file. See the supabase docs for more information.

1.2. Start Supabase locally

supabase start

1.3. Setup the Countdown

Clone the repository

git clone https://github.com/resonatehq-examples/example-openai-deep-research-agent-supabase-ts
cd example-openai-deep-research-agent-supabase-ts

Install dependencies

npm install

1.4. Serve the Countdown function

supabase functions serve deep-research-agent

1.5 Start the resonate server

resonate dev --system-url http://host.docker.internal:8001

1.6. Invoke the Deep Research Agent

Start a research task

resonate invoke <promise-id> --func research --arg <topic> --arg <depth> --target <function-url>

Example

resonate invoke research.1 --func research --arg "What are distributed systems" --arg 1 --target http://127.0.0.1:54321/functions/v1/deep-research-agent

1.7. Inspect the execution

Use the resonate tree command to visualize the research execution.

resonate tree research.1

research.1
├── research.1.0 🟢 (run)
├── research.1.1 🟡 (rpc research)
│   └── research.1.1.0 🟡 (run)
├── research.1.2 🟡 (rpc research)
│   └── research.1.2.0 🟡 (run)
└── research.1.3 🟡 (rpc research)
    └── research.1.3.0 🟡 (run)

You can also see the result with resonate promises get:

resonate promises get research.1

Supabase

Install the Supabase CLI

2.1 Deploy your secrets

supabase secrets set --env-file supabase/functions/.env

2.1 Deploy your function

supabase functions deploy deep-research-agent --project-ref <PROJECT_ID>

2.2 Start the resonate worker behind ngrok

ngrok http 8001

resonate dev --system-url <ngrok-url>

Example

resonate dev --system-url  https://583ef7749990.ngrok-free.app

2.4 Invoke the Deep Research Agent

Start a research task

resonate invoke <promise-id> --func research --arg <topic> --arg <depth> --target <function-url>

Example

resonate invoke research.1 --func research --arg "What are distributed systems" --arg 1 --target https://<orgId>.supabase.co/functions/v1/deep-research-agent

2.5. Inspect the execution

Use the resonate tree command to visualize the countdown execution.

resonate tree research.1

Example output (while waiting on the second sleep):

research.1
├── research.1.0 🟢 (run)
├── research.1.1 🟢 (rpc research)
│   └── research.1.1.0 🟢 (run)
├── research.1.2 🟢 (rpc research)
│   └── research.1.2.0 🟢 (run)
├── research.1.3 🟡 (rpc research)
├── research.1.4 🟢 (rpc research)
│   └── research.1.4.0 🟢 (run)
└── research.1.5 🟢 (rpc research)
    └── research.1.5.0 🟢 (run)

You can also see the result with resonate promises get:

resonate promises get research.1

Troubleshooting

If you are still having trouble please open an issue.