Browser agents

Software that turns the web into an action space.

Red ocean · Measured classification

Browser agents: Cooling recently, still above last year

Search attention for “browser agent” is mixed. 90 reviewed open-source projects serve this task. Established open-source alternatives shape this category. Compare their strengths and find a concrete reason for users to choose your product.

Report dated 2026-10-14 · Method 2.0.0 · low evidence confidence

Measured search term: browser agent · Last 8 complete weeks vs previous 8

≥749 active projects match the published GitHub search scope.
The keyword “browser agent” has 104 complete weeks; 0% are reported as zero.

Open-source competition: ≥74 / 100 · Established alternatives
Open-source alternatives: 90
Direction basis: Recent search windows
Matching active GitHub projects: ≥749
Search-interest growth: -24% · Last 8 complete weeks vs previous 8
Search term and region: browser agent · Worldwide
Complete weekly observations: 104

Why this classification

Recent search attention is cooling while staying above last year's level. Both horizons provide context.
Median weekly search interest fell 24% across two consecutive eight-week windows.
≥749 active repositories match this search scope. Review their users and workflows to identify direct alternatives.
Four-week search change: -1%; thirteen-week change: -53%. These windows check the direction of the eight-week comparison.
Project roles: 90 direct alternatives, 0 adjacent projects, 6 resources, 240 awaiting review.
Competition pressure: 74–100/100; breadth 48, established alternatives 19, leading project strength 7.

Source evidence

GitHub repository search · topic:browser-agent fork:false archived:false stars:>=1 pushed:>=2025-10-14

GitHub repository search · "browser agent" in:name,description fork:false archived:false stars:>=1 pushed:>=2025-10-14

GitHub repository search · topic:browser-automation topic:ai-agents fork:false archived:false stars:>=1 pushed:>=2025-10-14

GitHub repository search · topic:browser-automation topic:ai-agent fork:false archived:false stars:>=1 pushed:>=2025-10-14

Collected 2026-10-14T23:46:41.594Z

Google Trends search interest · Collected 2026-10-14T23:46:41.594Z

Leading repositories

Repository	Stars	Description
browser-use/browser-use	117,992	Agents that use the browser.
CloakHQ/CloakBrowser	32,204	Stealth Chromium that passes every bot detection test. Drop-in Playwright replacement with source-level fingerprint patches. 50+ tests passed.
feder-cr/invisible_dots	32,050	Open-source, self-hosted alternative to OpenAI Dots, Grok Bot. Built to be undetectable by anti-bot systems.
jackwener/OpenCLI	30,278	Make Any Website into CLI & Use your logged-in browser by AI agent.
alibaba/page-agent	29,448	JavaScript in-page GUI agent. Control web interfaces with natural language.
browserbase/stagehand	25,669	The SDK to extract data and interact with any site on the web. Get started with Claude Code, Codex, Eve, Mastra, and more.
browser-use/browser-harness	18,434	Browser Harness \| Self-healing harness that enables LLMs to complete any task.
lexmount/moli	18,381	Best headless browser for AI agents. Lite, Fast, High-Compatibility. Built in Rust
citrolabs/ego-lite	17,392	The fastest browser for AI agents to run browser automation, built for sharing your logged-in browser state with your AI agents, like Codex or Claude Code, without disturbing you. Zero cost, zero config.
browser-use/web-ui	16,615	🖥️ Run AI Agent in your browser.

Limits of this result

Google Trends measures relative search attention. Assess customers, revenue and willingness to pay through direct research.
Supply covers active repositories matching the displayed GitHub queries. Broader competition research can include other repositories and commercial products.
This search returned a partial sample. The displayed count is a lower bound.
Search windows or related terms show mixed directions. The competition assessment remains tied to the observed alternatives; compare the source curves for search momentum.
Repository density and search attention provide complementary research signals. Assess competition and demand through user and product evidence.
Competition pressure is an operational index of observed open-source alternatives. Stars and forks indicate developer attention and reuse; user adoption and commercial products deserve separate research.
The displayed projects form a stars-ranked sample. Competition pressure is a lower bound; wider coverage can strengthen the assessment.
Some project roles await closer review. The pressure range includes their possible contribution.

Search interest measures attention. Customer demand needs user research; classification thresholds are published heuristics subject to empirical calibration.

Community requests

Feature: machine-readable output format

Discussion open

It would be nice to be able to postprocess results of Magnitude test. I have two use-cases in mind: 1. In CI: run all tests, then upload all videos of failing tests to a storage, discard the rest, and produce the HTML page with the test results and links to

magnitudedev/browser-agent · Posted: 2025-04-28 · Updated: 2025-04-28 · Collected: 2026-10-14

allow users to download async the models in the onboarding process

Discussion open

1 decision model 2 visual model

webbrain-one/webbrain · Posted: 2026-10-11 · Updated: 2026-10-11 · Collected: 2026-10-14

Allow Magnitude tests to leverage LLM's "tools" on `check()` prompts

Discussion open

How about letting the check() prompts use Tools (like, mcp servers) ? How hard would that be? The goal would be to use english prompts to do validations for non-UI operations too (say, API or DB calls, using mcp's for OpenAPI servers and external DBs) (may

magnitudedev/browser-agent · Posted: 2025-04-29 · Updated: 2025-06-15 · Collected: 2026-10-14

Docs: clarify scope fingerprint spoofing vs IP-reputation/velocity blocks (proxy support?)

Discussion open

Summary Fortress does what it claims at the fingerprint level, but in testing against a site that gates on IP reputation/velocity (eBay), a clean fingerprint alone did not sustain access. It would help to state this scope boundary in the README, and possibly

tiliondev/fortress · Posted: 2026-09-17 · Updated: 2026-09-17 · Collected: 2026-10-14

Enable private vulnerability reporting

Discussion open

What happened Hi, I have a security finding to share privately. Could you enable private vulnerability reporting on this repo? Thanks Steps to reproduce Hi, I have a security finding to share privately. Could you enable private vulnerability reporting

tiliondev/fortress · Posted: 2026-08-20 · Updated: 2026-08-20 · Collected: 2026-10-14

bug: Fortress doesn't run on Debian 13

Discussion open

What happened I’m trying to run a program that uses Fortress on a Debian 13 VM via Proxmox. I’m getting the following error. On a regular computer with Win11, the program works fine. Steps to reproduce with Fortress(headless=False, channel="stable") as f

tiliondev/fortress · Posted: 2026-08-08 · Updated: 2026-08-10 · Collected: 2026-10-14

Use and share the evidence

Permanent report · Markdown · JSON · PNG card

Use the open-source CLI and MCP server on GitHub