OpenBrowserBLOCK
OpenBrowser is a framework for intelligent browser automation. It combines direct CDP communication with a CodeAgent architecture, where the LLM writes Python code executed in a persistent namespace, to navigate, interact with, and extract information from web pages autonomously.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
OpenBrowser
AI browser automation, powered by code.
Cloud dashboard for saved cookies and scheduled tasks is available in the hosted version. Join the waitlist for early access: https://openbrowser.me :
https://github.com/user-attachments/assets/b17f97f3-f9f8-4707-8e39-abbbbe1a693b
Automating Walmart Product Scraping:
https://github.com/user-attachments/assets/c517c739-9199-47b0-bac7-c2c642a21094
OpenBrowserAI Automatic Flight Booking:
https://github.com/user-attachments/assets/632128f6-3d09-497f-9e7d-e29b9cb65e0f
OpenBrowserAI Automatic Form Filling:
https://github.com/user-attachments/assets/16f7ef1a-beb1-45e2-a733-9592536e0ef7
[](https://pypi.org/project/openbrowser-ai/) [](https://pepy.tech/projects/openbrowser-ai) [](https://www.python.org/downloads/) [](https://opensource.org/licenses/MIT) [](https://github.com/billy-enrizky/openbrowser-ai/actions) [](https://codecov.io/gh/billy-enrizky/openbrowser-ai)
AI-powered browser automation using CodeAgent and CDP (Chrome DevTools Protocol)
OpenBrowser is a framework for intelligent browser automation. It combines direct CDP communication with a CodeAgent architecture, where the LLM writes Python code executed in a persistent namespace, to navigate, interact with, and extract information from web pages aut
9a191bf229c0OBSERVED · 2026-10-06Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add openbrowser-ai -- uvx openbrowser-ai==0.1.54
Trust audit
BLOCKgrade F · trust 35/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | FAIL |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (10 observation(s))
- Network
- declared (7 observation(s))
- Shell
- declared (3 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
add_special_tokens=False, # Must match eval (no special tokens for ReFusion)
exec(compiled, self._namespace)
"http://169.254.169.254/latest/api/token",
"http://169.254.169.254/latest/meta-data/instance-id",
"http://169.254.169.254/latest/meta-data/placement/region",
TOKEN=$(curl -s -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 60")
INSTANCE_ID=$(curl -s "http://169.254.169.254/latest/meta-data/instance-id" -H "X-aws-ec2-metadata-token: $TOKEN")
Run as Administrator or with Developer Mode enabled:
extension.zip
.DS_Store
.DS_Store
.DS_Store
task29-audio.ogg
module = importlib.import_module(module_path)
timing_text += f'Average per element: {avg_per_element:.2f} μs\n'new_config.llm[llm_id] = LLMEntry(id=llm_id, default=True, model='gpt-4.1-mini', api_key='your-openai-api-key-here')
.DS_Store
.DS_Store
.DS_Store
.dxtignore
module = importlib.import_module(module_path)
__builtins__.__import__(name, *args, **kwargs)
" eval(name)\n"
}`, n = new Function(Object.keys(h), m)(...Object.values(h)), m = `methodCaller<(${u.map((y) => y.name)}) => ${o.name}>`, Fi(Object.defineProperty(n, "name", { value: m }));h=`return function (handle, methodName, destructorsRef, args) {\n ${h}\n }`;c=(new Function(Object.keys(k),h))(...Object.values(k));h=`methodCaller<(${e.map(l=>l.name)}) => ${d.name}>`;return Ug(ObjGates applied: instruction_override, no_behavioural_pass.
9a191bf229c0full audit observations/trust-audit/mcp-server/billy-enrizky__openbrowser.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-06 | 9a191bf229c0 | BLOCK | F | 35 | first audit |
Questions
What is the OpenBrowser MCP server?
OpenBrowser is a framework for intelligent browser automation. It combines direct CDP communication with a CodeAgent architecture, where the LLM writes Python code executed in a persistent namespace, to navigate, interact with, and extract information from web pages autonomously.
Is OpenBrowser safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (35/100) and found 8 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does OpenBrowser need?
It reads ANTHROPIC_API_KEY, AZURE_OPENAI_API_KEY, AZURE_OPENAI_KEY, BEARER_TOKEN, BROWSER_USE_API_KEY, CEREBRAS_API_KEY, DEEPSEEK_API_KEY, DISCORD_BOT_TOKEN, GEMINI_API_KEY, GMAIL_CLIENT_SECRET, GMAIL_REFRESH_TOKEN and GOOGLE_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does OpenBrowser run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as openbrowser-frontend at 0.1.0.
How current is this page?
The grade is for one exact copy of the source (9a191bf229c0), read on 2026-10-06. The repository is watched and re-audited when it changes.