An AgentFeeds campaign

Don't be a
meat proxy.

You have better things to do than read the internet to your agent.

Reading newsletters. Scrolling LinkedIn. Pasting somebody’s take into a chat. Give your agent the sources and a question it can investigate instead.

Opens AgentFeeds. Public collections and profile downloads need no account.

The familiar routine01
NewsletterSocial postRelease notes
YouRead. Filter. Copy. Paste.
Your agentWaiting for context
A more direct route

Sources Your agent

You set the question and check the result.

Your attention
is not an API.

A post says a model is better. Your task needs to know: better at what, compared with which version, and under which conditions?

AgentFeeds gives your agent a curated source list and specific comparison questions. It can check the original evidence for the decision in front of it. You decide what matters and review the answer.

See how the handoff works →

Give your agent
a place to start.

Choose a decision. Take this instruction to the agent you already use.

Your agent needs permission to read the web. Its tools, model access and schedule stay under your control.

Agent instructionPlain text
Inspect the sources ↗

Read the sources yourself whenever you want.

Different agents need different reading lists. Customize this reading profile on AgentFeeds →

What you can use today

A reading list.
With a job to do.

Browse curated collections. Read them as Markdown, RSS or JSON Feed. Build a custom profile and download its settings and reading guide for each agent.

These are source references and research questions, not a live personalized news stream. Your agent reads the originals. Downloading a profile does not start a hosted schedule.

Explore AgentFeeds collections ↗

For the agent you already run

Point it at the source.

Choose the decisions your agent needs to stay informed about.

Build your AgentFeeds profile ↗

No model key needed to browse or download a reading profile.