collie
Herdr mobile app for iPhone and Android. A self-hosted PWA to drive Claude Code, Pi, Codex and OpenCode in Herdr, tmux or zellij from your phone. Push alerts, no app store.
Runssource: GitHubhomepageTypeScriptMITcommit 2bec9cbd6411
TypeScript, MIT licensed. The project labels itself: agent orchestration, ai, ai agents, ai coding, ai tools, claude code, codex and herdr.
collie runs. An Argusic agent installed it in 10 minutes and hit 6 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
Collie 1.14.1 builds from source, all 3668 bridge tests, 1613 CLI tests, and 13289 web tests pass; the bridge serves a real tmux session on HTTP with snapshot returning bridge:connected and shell panes.
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.
- Bun not found in environment1 minute
- Node.js 18.19.1 too old for Vite 8.x (needs 20.19+)1 minute
- No multiplexer installed (tmux/herdr/zellij missing)1 minute
- Port 8787 occupied by a system-level proxy (OpenRouter)1 minute
- Vitest 4.x uses `styleText` from node:util (Node 21+) not available in Node 18
- Bridge couldn't find tmux binary automatically
- Install time
- 10 minutes
- Cold machine to finish
- 9 minutes
- Errors hit and fixed
- 6 hit, 6 fixed with no human help
- How the result was proved
- Bridge tests: `bun test ./bridge` 3668 pass, 0 fail. CLI tests: `bun test ./cli` 1613 pass, 0 fail. Web tests: `bunx --bun vitest run` 13289 pass, 33 expected fail. Shell tests: all 10 scripts pass (collie-ctl, collie-cli, check-payload-links, upstream-bun, check-tag, check-flake-lock, pkgbuild, refresh, crew-harness, playground-entry). Launch: `curl http://127.0.0.1:8791/api/snapshot` returned HTTP 200 with `{"bridge":"connected"}` and valid session data from real tmux. `curl http://127.0.0.1:8791/` returned HTTP 200 serving the PWA index.html.
- Model tokens used
- 104,018
- Exact commit tested
- 2bec9cbd6411
- 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 2, 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.
exited 143 in 0ms: Now let me look at the total time spent and produce the final report. exec /bin/sh -lc 'echo "Start time estimate: 04:16 UTC"; date +"Current time: %H:%M UTC"; echo "~10 minu… succeeded in 0ms: Start time estimate: 04:16 UTC Current time: 04:25 UTC ~10 minutes total Here's the final report. tokens used 104,018 Here's the final report.
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 6 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 MIT, as reported by its host.
- Measured 2 times, so the result is not a one-off.
What did not, or is not known
- Took 10 minutes to install, slower than the median of the 21 comparable projects Argusic has measured.
Of the 22 agent-orchestration projects Argusic has installed and timed, collie was the 15th fastest to reach a running state, and 15 of 22 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.
- A quick evaluation: it was running 10 minutes after the clone.
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-orchestrationaiai-agentsai-codingai-toolsclaude-codecodexherdrherdr-clientherdr-pluginmobile-clientmultiplexingpwaself-hostedtailscale
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/collie)Questions
- Does collie run?
- Yes. collie runs. Argusic installed and launched it on a clean machine in 10 minutes, hitting 6 errors on the way, and recorded the session.
- How did Argusic test collie?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 2bec9cbd6411. 2 attempts are recorded, and the full method is on the methodology page.
- What did it cost to test collie?
- The run that produced this verdict cost $0.09: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does collie take to install?
- 10 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 collie need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing collie?
- 6 things broke in the recorded run, and 6 were fixed without human help. Each one, and the time it cost, is listed on this page.
- How does collie compare with the alternatives?
- Of the 22 agent-orchestration projects Argusic has installed and timed, collie was the 15th fastest to reach a running state, and 15 of 22 reached one at all.
- Where is the evidence for collie?
- All 2 recorded runs are on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.