unbrowse
Unbrowse - Reverse engineer any website so that your agents can access its APIs directly
Runssource: GitHubRustApache-2.0commit 7bcf436b8ca8
Rust, Apache-2.0 licensed. The project labels itself: agents, agentskills, browser use, claude code plugin, clawdbot, clawdbot plugin, codex plugin and gemini cli extension.
unbrowse runs. An Argusic agent installed it in 78 minutes and hit 5 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
The repo builds and its tests pass; unbrowse-node serves the agent API on 127.0.0.1:7788 and the node learned and then directly called a real route (httpbin.org/json), with agent keys minted against a schema-correct in-container Secret Service mock and one test-only no-sandbox patch to passkey.rs.
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. 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.
- Linux secret-service absent: root key creation failed35 minutes
- bindgen: libclang not found (btls-sys build)10 minutes
- rustup/cargo absent; foundry absent8 minutes
- passkey tests: Chrome crashed without --no-sandbox in container5 minutes
- learn_shop tests failed: debug agent-browser not built2 minutes
- Install time
- 78 minutes
- Cold machine to finish
- 80 minutes
- Errors hit and fixed
- 5 hit, 5 fixed with no human help
- How the result was proved
- cargo test --workspace --locked in unbrowse-node: 382 passed 0 failed; cargo test --locked in unbrowse-browser/cli: 1355 passed 0 failed (3 timed solo); unb whoami returned agent identity; unb do --url https://httpbin.org/json returned how=learned then how=called with JSON
- Model tokens used
- 8,668,435
- Exact commit tested
- 7bcf436b8ca8
- 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
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
- No root, so system packages were unavailable; libclang (needed by bindgen for BoringSSL) was extr… ## Errors found and fixed 1. **libclang missing** → extracted `libclang-18`/`libLLVM-18` .debs; set `LIBCLANG_PATH`, `LD_LIBR… 2. **Linux Secret Service absent** → built a schema-correct `org.freedesktop.secrets` D-Bus mock (D… 3. **Passkey test harness launched Chrome without `--no-sandbox`** (crash in this container) → patc… 4. **learn_shop tests failed** until the debug browser binary was built — fixed by `cargo build` in… ## Verified working (real, tier 1) - `cargo test --workspace --locked` in `unbrowse-node`: **382 passed, 0 failed** (incl. learn_shop … - `cargo test --locked` in `unbrowse-browser/cli`: **1355 passed (1352 + 3 timed solo), 0 failed, 1… - Live: `unbrowse-node` served `127.0.0.1:7788`; created agent `my-agent` (token `unb_78d9...`); `u… ## Final state - The node, `unb`, and `agent-browser` binaries exist and run; the app's node+CLI work end to end a…
Replay 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.
- Recovered from all 5 errors 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 Apache-2.0, as reported by its host.
What did not, or is not known
- Took 78 minutes to install, slower than the median of the 35 comparable projects Argusic has measured.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
Of the 36 agents projects Argusic has installed and timed, unbrowse was the 36th fastest to reach a running state, and 23 of 36 reached one at all.
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)
agentsagentskillsbrowser-useclaude-code-pluginclawdbotclawdbot-plugincodex-plugingemini-cli-extensiongrok-botharintegrationmcp-servermodel-context-protocolmoltbotmuseopenclawopenclaw-skillsreverse-engineeringskills
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/unbrowse)Questions
- Does unbrowse run?
- Yes. unbrowse runs. Argusic installed and launched it on a clean machine in 78 minutes, hitting 5 errors on the way, and recorded the session.
- How did Argusic test unbrowse?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 7bcf436b8ca8. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test unbrowse?
- The run that produced this verdict cost $1.79: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does unbrowse take to install?
- 78 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 unbrowse need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing unbrowse?
- 5 things broke in the recorded run, and 5 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does unbrowse compare with the alternatives?
- Of the 36 agents projects Argusic has installed and timed, unbrowse was the 36th fastest to reach a running state, and 23 of 36 reached one at all.
- Where is the evidence for unbrowse?
- 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.