Turn any site or docs into a clean, always-fresh feed that agents can actually read.
Point it at a domain and it produces and continuously maintains clean machine-readable representations: llms.txt and llms-full.txt, a hosted MCP resource server, per-page markdown with stable anchors, structured changelog and version metadata, and an embeddings-ready index.
No account. No card. No sales call.
- Clean feeds generated from any site or docs
- Re-crawled and diffed when content changes
- A hosted MCP resource server
Get a working setup in about ten seconds
Pick authentication, region, client and limits. Copy a config and client snippet that run, with the latency they imply. Runs in your browser. No account, no card, no call.
# my-service — generated by AgentFeeder
# Copy to my-service.config.yaml. Every value below is live, not a placeholder.
apiVersion: v1
kind: Service
metadata:
name: my-service
region: us-east
spec:
endpoint: https://my-service.agentfeeder.com/v1
auth:
mode: apikey
credential: ${MY_SERVICE_TOKEN}
limits:
requestsPerMinute: 600
timeoutMs: 10000
maxRetries: 2
backoff: exponential
observability:
structuredLogs: true
tracing: w3c
replayRetentionDays: 7curl -sS https://my-service.agentfeeder.com/v1/run \
-H "X-API-Key: $TOKEN" \
-H "Content-Type: application/json" \
--max-time 10.0 \
-d '{"input": {"task": "hello"}}'import { Client } from "@my-service/sdk";
const client = new Client({
endpoint: "https://my-service.agentfeeder.com/v1",
auth: { mode: "apikey", token: process.env.TOKEN! },
timeoutMs: 10000,
maxRetries: 2,
});
const result = await client.run({ task: "hello" });
console.log(result.data);- API keys do not expire on their own. Set a rotation reminder now.
Take this with you
No email required. It is your result.
AgentFeeder — config · my-service · us-east · apikey · p50 25ms, p99 80ms · 1,500,000 requests/month at 600 rpm burst Run it yourself: https://agentfeeder.com
Control how AI represents your product
A feed layer that turns a website into clean, always-fresh representations an assistant or an agent can read without parsing.
llms.txt and llms-full.txt
Point it at a domain and it produces and maintains llms.txt and llms-full.txt for the whole site.
A hosted MCP resource server
Your content is served through a hosted MCP resource server, ready for agents to call.
Per-page markdown
Every page is available as clean markdown with stable anchors, without navigation or cookie banners.
Changelog and version metadata
Structured changelog and version metadata keep pricing, API versions and limits current.
An embeddings-ready index
Content is indexed ready for embeddings, so agent builders can use it as a source straight away.
Re-crawled on change
It re-crawls when your content changes, diffs the result and notifies you.
The problem
An AI assistant asked about a product answers from a stale crawl of a JavaScript-rendered marketing site and gets the pricing, the API version and the limits wrong.
The vendor has no way to correct it.
Meanwhile agent builders burn tokens scraping HTML full of navigation and cookie banners to extract three paragraphs of substance.
Who it is for
Developer-tools and SaaS companies
Vendors with documentation who want AI assistants to get their pricing, API version and limits right.
Content businesses
Publishers who want control over how AI represents what they publish.
Agent builders
Developers who need reliable sources and are tired of scraping HTML for three paragraphs of substance.
How it works
AgentFeeder, from the first step to the result.
- 1
Point it at your domain
Add the site or documentation you want represented.
- 2
Feeds are generated
llms.txt, per-page markdown, an MCP resource server and an embeddings-ready index are produced.
- 3
Kept fresh automatically
Changes trigger a re-crawl, a diff and a notification.
Questions people actually ask
- What does AgentFeeder produce?
- llms.txt and llms-full.txt, a hosted MCP resource server, per-page markdown with stable anchors, structured changelog and version metadata, and an embeddings-ready index.
- How does it stay current?
- It re-crawls your site when content changes, diffs the result and notifies you.
- Why would a vendor want this?
- Assistants often answer from stale crawls of JavaScript-rendered sites and get pricing, API versions and limits wrong. A maintained feed gives them an accurate source you control.
- Why would an agent builder want this?
- It is a source that does not need parsing, so you are not burning tokens on navigation and cookie banners.
- What does it cost?
- 39 dollars a month per domain for hosted, auto-refreshing feeds with change notifications.
- Is there a larger plan?
- Yes. 249 dollars a month adds multiple domains, private sources, MCP hosting and analytics on which agents consumed what.
- Why should I trust the config generator?
- The config is built from the options you pick, and it lists anything in it that will misbehave under load before you copy it.
Pricing
Priced per domain, with a multi-domain plan for private sources and agent analytics.
DomainRecommended
Hosted, auto-refreshing feeds for one domain.
- Auto-refreshing feeds
- Change notifications
Multi-domain
For vendors with several sites or private content.
- Multiple domains
- Private sources
- MCP hosting
- Analytics on which agents consumed what
Prices in USD.
Get a working setup in about ten seconds
Pick authentication, region, client and limits. Copy a config and client snippet that run, with the latency they imply. Runs in your browser. No account, no card, no call.
Open the free toolIt runs in your browser. AgentFeeder never sees your inputs.