OpenAgentd

Open-source desktop app for a local coding agent. Watch every tool call, diff, and terminal command. Bring your own models; sessions stay on your machine.

Runs with mockssource: GitHubhomepageTypeScriptApache-2.0commit d44ddffc5b21

TypeScript, Apache-2.0 licensed. The project labels itself: agents, ai, ai coding assistant, coding agent, desktop app, llm, local first and mcp.

OpenAgentd runs, with stand-ins for the services it depends on. An Argusic agent installed it in 19 minutes and hit 4 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

The openagentd 2.26.0 backend server and web UI build cleanly, the suite passes 4728/4728 (12 skipped) plus 3168 web tests, the server answers 200 on /api/health/ready and /api/agents and /api/agents/registry, and a real chat turn completed against a local OpenAI-compatible mock, saving the assistant replies into the session.

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 92 of 100 (the mean of this project's run scores).

At a glance

verdict
Runs with mocks
Argusic Score
92 / 100
cost of the verifying run
$0.27 (measured)
recorded runs
1
last tested
stars
217
forks
38
open issues
2
watchers
0
size
83 MB
created
last push

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.

Time lost to each failure while testing OpenAgentd, drawn from the recorded run
Every bar is a failure the agent recorded, with the time it cost. Drawn from this run, not from anywhere else.

What went wrong, in the agent's own words

The agent's own words, worst first, never rewritten into advice.

  • tests/services/test_observability_service.py::test_summary_cache_invalidates_when_span_files_change failed: observability _run_queries call_count was 4 instead of 5; root cause: cache signature (path,size,mtime_ns,inode) did not detect append-then-truncate-back rewrites nor reliable rotation, because overlayfs quantizes mtime to milliseconds8 minutes
  • uv, bun, and unzip/npm-global were missing; system Node 18 too old for the web build (rolldown imports node:util styleText); desktop install.sh requires apt+sudo which are unavailable in the container4 minutes
  • Fresh first-run server start (make run / uvicorn, APP_ENV=development) crashed: no such table: scheduled_task because auto-migration only runs in production4 minutes
  • First end-to-end chat attempt hit ConnectError 5x because the backgrounded mock LLM process was reaped when its exec session ended1 minute
Install time
19 minutes
Cold machine to finish
20 minutes
Errors hit and fixed
4 hit, 4 fixed with no human help
How the result was proved
uv run pytest -n 4 -rf exited 0: 4728 passed, 12 skipped; curl http://127.0.0.1:4082/api/health/ready returned 200 {"status":"ok","version":"2.26.0","checks":{"db":"ok","agent":"ok"}}; bun test --parallel (web) exited 0: 3168 pass 0 fail; POST /api/agent/chat with workspace /tmp/ws against a local OpenAI-compatible mock (OLLAMA_BASE_URL) returned 202 accepted and the session history showed assistant 'Mock reply to: Say hello' and 'Mock reply to: hello again'
Model tokens used
1,463,517
Exact commit tested
d44ddffc5b21
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
Behaviour against real services
this run was verified against stand-in services only

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.

- Full backend suite: `4728 passed, 12 skipped` (was `4727 passed, 1 failed` before my fix).
- Web: lint clean, typecheck clean, `3168 pass, 0 fail`; manual service scenarios 21/25/30/39 all p…
- **Error found & fixed**: `tests/services/test_observability_service.py::test_summary_cache_invali…
- **Real end-to-end agent turn**: stood up a schema-correct Argusic-compatible mock (Ollama `/v1/ch…
- Only gap: native Rust shells (desktop/mobile/shell-core) can't be checked — no `cargo`/Rust toolc…
- Elapsed: ~19 minutes.
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, "errors": [{"msg": "uv…
tokens used
1,463,517

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

  • Recovered from all 4 errors without a human stepping in, which says the failures are documented well enough to solve.
  • Licensed Apache-2.0, as reported by its host.

What did not, or is not known

  • Only reached a running state against stand-in services, so nothing here proves it works against the real ones.
  • Took 19 minutes to install, slower than the median of the 40 comparable projects Argusic has measured.
  • Measured once so far. A single attempt shows it can run, not that it runs reliably.

Of the 41 agents projects Argusic has installed and timed, OpenAgentd was the 36th fastest to reach a running state, and 29 of 41 reached one at all.

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

tested byattemptoutcomescorecost USDwhen
Argusic Runner1/3Runs with mocks92.000.27

Topics (from GitHub)

agentsaiai-coding-assistantcoding-agentdesktop-appllmlocal-firstmcpmulti-agentollamarustself-hostedtauri

Embed the badge

Markdown for the project README. It links back here; terms on the terms page.

[![Tested by Argusic](https://argusic.com/badge/OpenAgentd.svg)](https://argusic.com/subject/openagentd)

Questions

Does OpenAgentd run?
OpenAgentd runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 19 minutes, hitting 4 errors on the way, and recorded the session.
How did Argusic test OpenAgentd?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit d44ddffc5b21. One attempt is recorded, and the full method is on the methodology page.
What did it cost to test OpenAgentd?
The run that produced this verdict cost $0.27: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does OpenAgentd take to install?
19 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 OpenAgentd need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing OpenAgentd?
4 things broke in the recorded run, and 4 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does OpenAgentd compare with the alternatives?
Of the 41 agents projects Argusic has installed and timed, OpenAgentd was the 36th fastest to reach a running state, and 29 of 41 reached one at all.
Where is the evidence for OpenAgentd?
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.

Discussion