cd /news/ai-agents/agents-in-action-5-real-ai-agents-on… · home › topics › ai-agents › article
[ARTICLE · art-170718] src=blog.postman.com ↗ pub= topic=ai-agents verified=true sentiment=↑ positive

Agents in Action: 5 Real AI Agents on Astropods

Postman's Astropods agent platform launched a video series, "Agents in Action," profiling five production agents built on the Astropods agent operating system layer, including The Reading Room, Ada, Memory Box, Cruise Line and an Industry News Agent. Astropods runs each agent through a project with an astropods.yml spec, a versioned blueprint and a deployed agent with observability covering traces, evals, token usage, spend and latency, and the blueprints are available for cloning. Ada, a cloud-native autonomous coding agent co-created by Rodric Rabbah, routes models through the Astropods AI Gateway so builders can swap between Claude, Kimi K3 and self-hosted open-weights or on-prem models.

by read7 min views1 publishedNov 16, 2026
Agents in Action: 5 Real AI Agents on Astropods
Image: Blog (auto-discovered)

Most agent demos end the moment the agent answers a question. You never see how it was wired up, what it talks to, or whether it would hold up in production. That’s the part worth showing.

Agents in Action is a video series that does exactly that. Each episode opens up a real agent running on Astropods: the running system, the config behind it, and a short conversation with the person who built it about the choices they made. No slides, no toy demos. Here’s the lineup, and the patterns worth borrowing for your own agents.

The lineup at a glance #

Agent What it does Pattern to steal
The Reading Room Turns newsletters in Gmail into a browsable library of books A custom frontend instead of a chat box
Ada Cloud-sandboxed autonomous coding agent Model routing through an AI Gateway, shared team context
Memory Box Natural-language search over your bookmarks and saved items Vector search in a sandboxed, single-purpose agent
Cruise Line Analyzes PRs; you pick which comments get posted Human-in-the-loop plus team-specific rule sets
Industry News Agent Posts a daily tech news briefing when you @mention it Parallel API fan-out, dedupe, deploy to Slack

Astropods in 60 seconds #

Astropods is the agent operating system layer in the Postman AI stack. It runs and controls agents in production, with runtime, observability, and guardrails built in.

Every agent here follows the same lifecycle: a project (your code plus an astropods.yml spec for models, interfaces, integrations, knowledge stores, and runtime needs), published as a versioned blueprint, deployed as a running agent, and watched through observability (traces, evals, token usage, spend, latency).

ast project create my-agent
ast deploy

Because the same spec drives local dev and production, what you watch in these episodes is what you can run yourself. The full walkthrough is in Build, Deploy & Monitor AI Agents with Astropods.

1. The Reading Room: an agent with its own UI #

Newsletter overload, solved with an interface that isn’t a chat box. In this episode I sit down with Taylor Green, Founding Product Designer at Astropods, to tour The Reading Room: a library entrance, digital bookshelves, and newsletters rendered as books you read chapter by chapter.

Under the hood

  • Pulls newsletters from Gmail using a label and an app password.
  • Uses Anthropic Claude to extract key insights, highlights, and full article text.
  • Stores everything in a serverless Neon database.
  • Astropods hosts the custom web frontend and auto-provisions pods, domains, and hosting, so there’s no infrastructure to manage.
  • Digests run manually or on a schedule, with custom ingest animations.

Steal this: if your users need to browse, compare, or reread what an agent produced, give it a purpose-built UI. Clone the blueprint and point it at your own inbox.

Blueprint: The Reading Room

2. Ada: a coding agent that lives in the cloud #

What if your coding agent wasn’t trapped on your laptop, locked to one model provider, or cut off from the rest of your engineering team? I sit down with Rodric Rabbah, Head of Product and co-creator of Astropods, to break down Ada, a cloud-native autonomous coding agent built directly on the Astropods platform and named after Ada Lovelace. It’s a good example of a custom agent deployed in production.

Under the hood

  • Runs in a secure, sandboxed terminal in the cloud, so it isn’t tethered to your laptop.
  • Two ways to talk to it: a custom terminal frontend built on AG-UI, or the built-in Astropods Chat playground.
  • The Astropods AI Gateway routes models, so you can swap between Claude, Kimi K3, and self-hosted models, including open-weights and on-prem options.
  • “Shared intelligence” compounds across the whole dev team. Rodric explains why with a central heating and plumbing analogy.
  • Persistent memory lives in PostgreSQL, with access controls and identity management.
  • Deploy it from the Blueprint Registry, with automated GitHub CI/CD syncing.
  • The team used Ada to build and iterate on Ada itself.

Use this: to route models through a gateway so you’re never locked to one provider, and keep an agent’s context somewhere the whole team can benefit from it.

Blueprint: Ada: Your Personal Coding Agent

3. Memory Box: search everything you saved #

That X bookmark you can’t find. The GitHub star you saved months ago and can’t name. Memory Box, built by community member Christopher Patty, consolidates your saved items into one place and lets you find them by asking.

Under the hood

  • Vectorization powers natural-language search across all your saved content.
  • A chat interface handles retrieval.
  • Every deployment is fully sandboxed, so your private data stays private.
  • Setup needs an Anthropic key and a few steps.

Use this: as a single-purpose agent with a narrow job is easy to build, easy to reason about, and easy to deploy.

4. Cruise Line: code review with you in the loop #

AI is generating more code, PRs are piling up, and many review tools respond by flooding them with noisy comments. Cruise Line flips that: AI analyzes the PR, and you decide which comments are worth posting to GitHub.

Under the hood

  • Automated analysis covers security, performance, code reuse, and maintainability.
  • Custom rule sets enforce team-specific standards, such as blocking plain string SQL injections.
  • A built-in chat lets you ask questions about the code without pulling it down locally.
  • Deploy by connecting your GitHub repo and adding an Anthropic key.
  • Published as a public blueprint, so any team can deploy its own instance.

Use this: to keep a human in the loop for any action that’s visible to others. The agent proposes, you approve.

Blueprint: Cruise Line

5. Industry News agent: a daily briefing in Slack #

This one is a good template for deploying any agent to Slack. I sit down with Astropods Software Engineer Sohum Dalal to deploy a public blueprint that gives anyone on your team a newsletter-style briefing from a single @mention.

Under the hood

  • Queries NewsAPI, GNews, The Guardian, and MediaStack in parallel, then removes duplicate articles.
  • Found and inspected through the public blueprints in the Astropods Explore registry.
  • Required and optional API keys are set in the deployment form.
  • Slack setup: an app-level token with connections:write , Socket Mode, Event Subscriptions withapp_mentions:read , and bot token scopes for channel history, posting, and direct messages.

Use this: to fan out to several sources at once and dedupe before the model sees anything. Then put the agent where your team already works.

Blueprint: Industry News Agent

Patterns that show up across the series #

  • Narrow beats broad. Memory Box, Cruise Line, and the news agent each do one job well.
  • The interface is part of the agent. Chat, a custom UI, a cloud terminal, and Slack each fit a different job.
  • Don’t lock yourself in. Ada routes models through the AI Gateway, so swapping providers is a config change, not a rewrite.
  • Keep humans in charge of visible actions. Cruise Line lets you approve before anything posts.
  • Ship it, then watch it. Traces, evals, and spend monitoring come with the platform rather than being bolted on later.

Run one yourself in 10 minutes #

  1. Install the ast CLI and start Docker (seeinstall the CLI ).
  2. Browse the Blueprint Registry at astropods.com , pick an agent from the series, and deploy it with your own keys. Or runast project create and build your own.
  3. Connect it to Slack with --adapter slack , then check its traces and token usage once it’s live.

Prefer a guided path? Try the Building with Agents learning path, or start with Your first project. Already have an agent? The Astropods Claude Code plugin can help migrate it.

Watch the series and join in #

Binge the full Agents in Action playlist, and join the community on Discord. Built something you’d like to see in the series? Share it there and it might be next.

── more in #ai-agents 4 stories · sorted by recency
── more on @astropods 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/agents-in-action-5-r…] indexed:0 read:7min 2026-11-16 · —