agent-chrome-relay
Let your AI agents use your Chrome, signed in as you, in background tabs. By Alexey Indeev · blog: read.aindeev.com · X: @AlexeyIndeev
Runssource: GitHubhomepageGoMITcommit ca2f952c04b4
Go, MIT licensed. The project labels itself: agent browser, ai agents, browser automation, cdp, chrome extension, claude, claude code and codex.
agent-chrome-relay runs. An Argusic agent installed it in 19.2 minutes and hit 1 error and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
The Chrome Relay relay server starts, answers HTTP endpoints with correct status codes, and passes all 12 integration tests including agent WebSocket connections authenticated by secret and process markers, extension origin validation, web page rejection, and audit trail persistence.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 100 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- lsof is not available on Linux; relay's caller() function could not find connecting process PIDs, causing all agent connections to be rejected with code 4003 (tests 4-7, 12 failed)18 minutes
- Install time
- 19 minutes
- Cold machine to finish
- 20 minutes
- Errors hit and fixed
- 1 hit, 1 fixed with no human help
- How the result was proved
- cd relay && npm test exited 0, 12/12 passed including agent WebSocket auth, CDP command dispatch, extension origin checks, web page rejection, and audit trail recording. curl http://127.0.0.1:<port>/status returned 200 with profiles and agents arrays. curl http://127.0.0.1:<port>/json/version returned 200 with correct CDP metadata.
- Model tokens used
- 488,086
- Exact commit tested
- ca2f952c04b4
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
Watch the session
The whole run, from a clean machine, at the speed it actually happened. Nothing is cut and nothing is sped up.
A recording of attempt 1, run , the attempt that decided the verdict above. The terminal recording it was made from is the evidence, stored with a sha256 fingerprint; the video is a rendering of it. Nothing loads from the video host until you press play.
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
Start time: 1793031374
Current time: 1793032526
**Start:** 1793031374
**End:** 1793032526
**Duration:** ~19.2 minutes
All observations complete. Here is the final report:
exec
/bin/sh -lc "echo '{\"install_succeeded\": true, \"launch_succeeded\": true, \"install_minutes\": 1…
succeeded in 0ms:
{"install_succeeded": true, "launch_succeeded": true, "install_minutes": 19.2, "errors": [{"msg": "…
tokens used
488,086Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 19.2 minutes.
- Recovered from all 1 error without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
Topics (from GitHub)
agent-browserai-agentsbrowser-automationcdpchrome-extensionclaudeclaude-codecodexcursormacospaseo
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/agent-chrome-relay)Questions
- Does agent-chrome-relay run?
- Yes. agent-chrome-relay runs. Argusic installed and launched it on a clean machine in 19 minutes, hitting 1 error on the way, and recorded the session.
- How did Argusic test agent-chrome-relay?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit ca2f952c04b4. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test agent-chrome-relay?
- The run that produced this verdict cost $0.10: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does agent-chrome-relay take to install?
- 19.2 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does agent-chrome-relay need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing agent-chrome-relay?
- 1 thing broke in the recorded run, and 1 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for agent-chrome-relay?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.