[Submission] Daily News Briefing Agent

[melienherrera]

Project link

https://github.com/temporal-community/daily-news-briefing-agent-demo

Language

TypeScript

Short description (max 256 chars)

A demo comparing AI agent workflows with and without durable execution. Generates daily tech briefings via web search, summarization, and AI synthesis — showing how Temporal handles retries and crash recovery automatically.

Long Description

What is this?

This project demonstrates what it looks like to build an AI agent that actually works in production — not just in a happy-path demo. It runs a daily tech briefing agent that searches the web for topics you
care about, summarizes the findings, and synthesizes everything into a personalized report.

The twist: it runs the same workflow twice, simultaneously — once as a plain async implementation, and once powered by Temporal. The side-by-side UI makes the difference immediately
visible.

---

The problem with AI agents in production

Most AI agent demos work great — until they don't. A network timeout mid-workflow, an API rate limit after step 3 of 5, a process crash while waiting on a slow LLM call. In a vanilla implementation, any of these
means starting over from scratch. In long-running, multi-step agents, that's expensive and frustrating.

Temporal solves this by making workflows durable by default. Every step is checkpointed. If anything fails, it retries automatically with configurable backoff. If the process crashes entirely, it resumes
exactly where it left off when it comes back up.

---

How the agent works

The briefing agent runs in three phases:

Topics entered by user
│
▼
┌───────────────────┐
│ Phase 1: Search │ ← Parallel web searches per topic
└───────────────────┘
│
▼
┌─────────────────────┐
│ Phase 2: Summarize │ ← AI summarizes each topic's findings
└─────────────────────┘
│
▼
┌──────────────────────┐
│ Phase 3: Brief │ ← Final report synthesized from all summaries
└──────────────────────┘

Both implementations follow the same logic. The difference is what happens when something goes wrong.

---

Side-by-side comparison

The web UI runs both versions at the same time. You can watch the phase-by-phase progress live and see how each handles failures.

	Without Temporal	With Temporal
Phase 1 failure	Job errors out, all progress lost	Retries automatically with backoff
Process crash	All in-flight work is gone	Resumes from last checkpoint on restart
Observability	Terminal logs only	Full workflow history in Temporal UI
Long-running steps	Fragile	Fault-tolerant
Retry logic	Write it yourself	Built-in, configurable per activity

To simulate a crash on the vanilla side: just kill the server (Ctrl+C) while a job is running. Restart it and the vanilla job is gone. The Temporal workflow will pick back up automatically once the worker
reconnects.

---

Demo

📹 https://www.youtube.com/watch?v=zd4g_ZrMj6A

The Temporal column shows:

• Live phase progress as each step completes
• A retry badge when Phase 1 is automatically retried
• A direct link to the Temporal UI to inspect workflow history, event timeline, and activity results

---

Tech stack

• Temporal — durable workflow execution
• AI SDK — LLM calls and tool use
• OpenAI — gpt-4o-mini with built-in web search
• TypeScript — end to end
• Express — lightweight server for the web UI

---

Key concepts illustrated

• Activities vs. Workflows — non-deterministic work (API calls, web search) lives in activities; orchestration logic lives in workflows
• Automatic retries — Phase 1 intentionally fails on the first attempt to show Temporal retrying transparently
• Workflow queries — the UI polls live phase progress directly from the running workflow via a Temporal query handler
• Durable state — workflow progress survives worker restarts with no extra code

Generated by Claude

Author(s)

Melissa Herrera

[github-actions]

Hi, I'm ZiggyBot! 🤖 Here's my pre-evaluation of this submission:

---

Criterion	Assessment
Useful to Temporal users	✅ Yes — This project provides an excellent educational comparison showing the concrete benefits of durable execution for AI agent workflows, which is highly relevant given the growing interest in LLM-based applications.
Benefits clearly explained	✅ Yes — The documentation clearly articulates the problem (AI agents failing mid-workflow) and demonstrates Temporal's solution through a side-by-side comparison with detailed tables and workflow diagrams.
OSI Approved License	⚠️ Unclear — GitHub did not detect a license file in the repository, which needs to be addressed before acceptance.
README with working instructions	✅ Yes — The README provides clear step-by-step setup instructions including prerequisites, terminal commands for all three required processes, and both web UI and CLI usage options.

Overall: Needs review ⚠️

Notes:

• This is a well-crafted educational demo that effectively showcases Temporal's value proposition for AI agent reliability — the side-by-side comparison approach is particularly compelling for newcomers.
• The project demonstrates integration with @temporalio/ai-sdk and the Vercel AI SDK, which is a timely and practical combination.
• The inclusion of a YouTube demo video is a nice touch for learners who prefer visual content.
• The repo is a fork with 0 stars, which is fine for a new submission, but the missing license is a blocker.
• The workflow code shows clean separation of concerns with Activities handling non-deterministic LLM calls while the Workflow handles orchestration.

Suggested questions for submitter:

• Could you add an OSI-approved license file (MIT recommended) to the repository?
• Is temporal-community the intended long-term home for this project, or will it move to a personal/organization account?

Teaching moment:

This project demonstrates an excellent pattern for integrating AI SDK tool calls with Temporal Activities. In hello-world-workflow.ts, notice how the getWeather and calculateAreaOfCircle functions are first wrapped with proxyActivities (giving them Temporal's retry and durability guarantees), and then passed directly to the AI SDK's tool() definitions. This means when the LLM decides to call a tool, that execution automatically benefits from Temporal's fault tolerance — a powerful pattern for production AI agents where tool calls might hit flaky external APIs.
View in source ↗

---

---

ZiggyBot is an AI pre-screener based on Temporal's community mascot Ziggy. Final decisions are made by the community team.

Reviewer: Check the box below to generate a Contentful-ready long description for this submission.

• Generate long description if needed

[github-actions]

Hi, I'm ZiggyBot! 🤖 Here's my pre-evaluation of this submission:

---

Criterion	Assessment
Useful to Temporal users	✅ Yes — Demonstrates a practical side-by-side comparison of durable vs. non-durable AI agent execution, which is highly valuable for developers evaluating Temporal for LLM-based workflows.
Benefits clearly explained	✅ Yes — The documentation clearly articulates crash recovery, automatic retries, and observability advantages through comparison tables and detailed workflow explanations.
OSI Approved License	⚠️ Unclear — GitHub did not detect a license file in the repository, which needs to be addressed before acceptance.
README with working instructions	✅ Yes — Instructions are well-structured with clear prerequisites, multiple terminals setup, and both CLI and web UI options documented.

Overall: Needs review ⚠️

Notes:

• This is a thoughtfully designed educational demo that showcases Temporal's value proposition for AI agent workflows in a compelling way
• The side-by-side comparison approach with intentional failure injection is an excellent teaching tool
• The project integrates @temporalio/ai-sdk with Vercel AI SDK, demonstrating a real-world integration pattern
• The repo is a fork with 0 stars, which is fine for a demo but worth noting
• The video demo linked in the description adds significant value for understanding the project
• Missing license is the only blocking issue — once added, this would be a strong candidate for the Code Exchange

Suggested questions for submitter:

• Could you add an OSI-approved license (MIT recommended) to the repository?
• Is the @temporalio/ai-sdk package publicly available, or is this demo tied to a specific preview/beta release?
• Would you consider adding a brief section on how users could extend this pattern to their own AI agent use cases?

Teaching moment:

This project demonstrates an elegant pattern for wrapping LLM tools as Temporal Activities, keeping non-deterministic API calls (web search, AI generation) isolated from the deterministic workflow orchestration. The tool({ execute: getWeather }) pattern shows how existing AI SDK tools can be backed by proxied activities, giving you automatic retries and durability without changing your tool interface. This is a great example of how Temporal's activity model fits naturally with modern AI agent tooling patterns.
View in source ↗

---

---

ZiggyBot is an AI pre-screener based on Temporal's community mascot Ziggy. Final decisions are made by the community team.

Reviewer: Check the box below to generate a Contentful-ready long description for this submission.

• Generate long description if needed

[webchick]

Added! https://temporal.io/code-exchange/daily-news-briefing-agent-demo-vercel-ai-sdk