---
type: "post"
id: "relinos-organizing-the-web-133f22"
slug: "relinos-organizing-the-web-133f22"
title: "relinos: Organizing the web"
description: "Relinos is an open-source programmable browser exposed through a REST API for rendering pages, interacting with sites and extracting structured data."
retrieval_nugget: "Relinos is an open-source programmable browser exposed through a REST API for rendering pages, interacting with sites and extracting structured data. Its routines can make browser workflows reusable, giving builders a self-hosted component for web automation rather than a managed service."
published_at: "2026-11-02"
updated_at: "2026-11-02"
record_date: "2026-11-02"
date_kind: "published_at"
topics: ["agent-tools","computer-use","automation"]
entities: ["Relinos"]
related_patterns: ["agent-ready-software-exposes-capabilities"]
novelty_label: "new_fact"
what_changed: "Relinos is a distinct programmable browser for automating web browsing, unlike the prior browser-automation and browser project reports."
prior_refs: ["https://newruntime.com/posts/northstar-browser-northstar-web-browser-open-source-license-a149ce/"]
source_url: "https://github.com/realsirjoe/relinos"
source_urls: ["https://github.com/realsirjoe/relinos"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:relinos-organizing-the-web-133f22"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/relinos-organizing-the-web-133f22/","markdown":"https://newruntime.com/posts/relinos-organizing-the-web-133f22.md","json":"https://newruntime.com/posts/relinos-organizing-the-web-133f22.json"}
markdown_url: "https://newruntime.com/posts/relinos-organizing-the-web-133f22.md"
---

# relinos: Organizing the web

## Retrieval answer

Relinos is an open-source programmable browser exposed through a REST API for rendering pages, interacting with sites and extracting structured data. Its routines can make browser workflows reusable, giving builders a self-hosted component for web automation rather than a managed service.

## Summary
Relinos is an open-source programmable browser exposed through a REST API for rendering pages, interacting with sites and extracting structured data. Its routines can make browser workflows reusable, giving builders a self-hosted component for web automation rather than a managed service.

## What happened
The public GitHub repository describes Relinos as “a programmable browser that does the browsing for you.” It provides a REST API for rendering and interacting with websites, controlling multiple browsers and extracting structured data. The documented setup runs on Linux and macOS: clone the repository, check out v0.1.0, install it with the included Python installer, and start the service with a shell script. The README’s example calls the local service at port 8032.

The central `/get` endpoint accepts a URL and options. A caller can request browser rendering, choose an output format—HTML, Markdown or JSON—and set a cache duration. Other parameters include a proxy, a debug response header, metadata wrapping, pagination and extraction rules. Rendering is handled with a Chromium browser such as Google Chrome when the `render` parameter is enabled, which lets the service handle sites that depend on JavaScript.

For interaction, a request can include an ordered list of actions. The documented actions include waiting for and selecting elements by XPath, clicking, typing, pressing Enter, navigating to another URL, executing JavaScript, sleeping, and filtering selected elements by text or attributes. The README’s example demonstrates a login flow; another searches a site by entering a query and submitting it. These are examples of interface mechanics, not evidence that the tool can safely or reliably complete arbitrary workflows.

Relinos also lets users save a sequence as a named routine through a `/routines/<name>` endpoint. Routines can accept parameters and patch them into request fields, with options such as URL quoting and typed values. Existing routines can be overwritten using a query parameter. The repository recommends multiple routines for complex tasks, with search and purchasing as examples. A registry section lists ready-to-use interactions and extractions for services including Google Search, Google Maps, YouTube, DuckDuckGo, eBay and Skyscanner; the README says these can be imported through a `/import` endpoint. The repository is public and shows 28 stars and two forks. The supplied material gives no hosted-service offer, price, service-level commitment or quantified performance claim.

## What the archive adds
[Northstar-browser](https://newruntime.com/posts/northstar-browser-northstar-web-browser-open-source-license-a149ce/) covered an open-source browser-engine project with deliberately narrow scope. Relinos is a new event in the adjacent browser-automation thread, not evidence that it shares Northstar’s architecture. [Mantis](https://newruntime.com/posts/mantis-an-open-source-mcp-browser-interface-claims-7592df/) and [Stagehand v4](https://newruntime.com/posts/stagehand-v4-moves-browser-agent-state-and-policy-closer-to-the-browser/) are related browser-agent stories, but the available archive summaries do not establish direct technical links.

## Trend signal
This is a modest example of [agent-ready software exposing capabilities](https://newruntime.com/patterns/agent-ready-software-exposes-capabilities/): Relinos exposes browser actions and extraction through explicit API inputs. The source does not describe a permissions model, external-agent integration or safety controls, so it supports the pattern only at the level of callable capabilities, not the broader claim that they are safely composable.

## Why it matters
A browser is often the practical bridge between an agent and services that lack an API or require interaction with a web interface. Relinos packages rendering, interaction and extraction behind a local HTTP interface, which can help a team prototype or operate workflows without embedding browser-control code in every agent. Named, parameterized routines offer a way to separate a reusable workflow from the inputs that vary between runs.

That convenience shifts responsibility to the operator. The README’s examples include credentials and actions such as clicking and typing, while the supplied material does not explain secret handling, access control, isolation between browser sessions or guardrails against consequential actions. Teams evaluating the project should treat it as an automation component to inspect and secure, not as a turnkey agent runtime. Its small visible adoption footprint also makes maintenance and compatibility worth validating before relying on it in production.

## Limits and uncertainty
The repository documents setup, endpoints and example interactions, but the supplied material does not establish how actively the project is maintained, which Chromium versions it supports, or how it behaves across complex sites. It offers no benchmark, uptime, scale or reliability data. The README mentions controlling multiple browsers but does not specify concurrency limits or isolation guarantees. Nor does it explain authentication, credential storage, permission boundaries or safeguards for purchases and other high-impact actions. The presence of sample registry routines does not establish that those integrations remain current or work consistently.

## Skill to build now
Take one low-risk, repeatable web task and write down its inputs, expected page state and verifiable output before automating it. If evaluating Relinos, begin with a public site and a read-only workflow: render a page, select a known element, extract a field, and check the result against the page manually. This gives a concrete test of the API without exposing credentials or authorizing irreversible actions.

## Reader takeaway
Relinos offers a self-hosted REST interface for browser rendering, actions, reusable routines and extraction. It is a useful project to inspect when building web automation, but the repository’s examples are not proof of production readiness. Validate compatibility and build your own security boundaries before connecting it to an agent or sensitive workflow.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Agent Tools](https://newruntime.com/topics/agent-tools.md): Explore the agent-tools topic hub.
- topic hub: [Computer Use](https://newruntime.com/topics/computer-use.md): Explore the computer-use topic hub.
- topic hub: [Automation](https://newruntime.com/topics/automation.md): Explore the automation topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Sources](https://newruntime.com/sources/): See the original sources this archive cites.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)