swarmclaw

Open-source self-hosted AI agent runtime and multi-agent framework for autonomous agent swarms. Agent memory, MCP tools, schedules, delegation, and 23+ LLM providers (Claude, GPT, Gemini, OpenRouter, Ollama). A practical Claude Code and LangChain alternative.

Runssource: GitHubhomepageTypeScriptMITcommit ed38ba5329c2

TypeScript, MIT licensed. The project labels itself: agent framework, agent memory, agent runtime, agent swarm, agents, ai, ai agent framework and ai agents.

swarmclaw runs. An Argusic agent installed it in 43 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.

SwarmClaw installs, builds, passes all 938 tests, and runs: the built and dev servers both answer 200 on /api/healthz, the web app serves HTML on /, the CLI answers --help and version 1.9.40, and the Electron desktop app launches on the virtual display and loads its UI.

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

verdict
Runs
Argusic Score
100 / 100
cost of the verifying run
$0.06 (measured)
recorded runs
1
last tested
stars
712
forks
143
open issues
25
watchers
9
size
15 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 swarmclaw, 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.

  • 127.0.0.1 loopback refused connections inside the container (container-level network namespace allows only the container IP)8 minutes
  • System Node was v18.19.1 but the project requires Node >=22.6, and setup.hard-fails on older Node7 minutes
  • Electron main process launches but tried to load http://127.0.0.1:3456 and initially hit ERR_CONNECTION_REFUSED5 minutes
Install time
43 minutes
Cold machine to finish
9 minutes
Errors hit and fixed
3 hit, 3 fixed with no human help
How the result was proved
npm test exited 0 (938 pass, 0 fail); curl -s -o /dev/null -w %{http_code} http://172.19.0.30:3000/api/healthz returned 200 and / returned 200; electron electron-dist/main.js on DISPLAY=:99 stayed up with did-finish-load
Model tokens used
208,205
Exact commit tested
ed38ba5329c2
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.

**Findings**
- System Node is v18.19.1, but the repo requires Node ≥22.6 (`.nvmrc`, README, and `easy-setup.mjs`…
- Container networking: loopback `127.0.0.1` refuses connections here and the Next server binds all…
**Verification performed**
- `npm install` → exit 0, 2002 packages added; postinstall reported no errors.
- `npm test` → exit 0, aggregate 938 tests across suites: 95+160+160+481+42 all pass, 0 fail, 0 ski…
- `npm run build` → exit 0, `.next/standalone` produced.
- Production server (`node .next/standalone/server.js`): `GET http://172.19.0.30:3000/api/healthz` …
- Dev server (`npm run dev`, turbopack): `GET http://172.19.0.30:3456/api/healthz` → 200.
- CLI: `node bin/swarmclaw.js --help` exited 0 with usage text; `node bin/swarmclaw.js version` pri…
- Desktop app: `electron electron-dist/main.js --no-sandbox` on DISPLAY=:99 → main + renderer proce…
- Unchecked gap: one screenshot-bearing page requires a `PHANTOM_TOKEN` env secret for a third-part…

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 3 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.

What did not, or is not known

  • Took 43 minutes to install, slower than the median of the 63 comparable projects Argusic has measured.
  • Measured once so far. A single attempt shows it can run, not that it runs reliably.

Of the 64 agent-framework projects Argusic has installed and timed, swarmclaw was the 63rd fastest to reach a running state, and 51 of 64 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

tested byattemptoutcomescorecost USDwhen
Argusic Runner1/3Runs100.000.06

Topics (from GitHub)

agent-frameworkagent-memoryagent-runtimeagent-swarmagentsaiai-agent-frameworkai-agentsautonomous-agentsclaudeclaude-code-alternativelangchain-alternativellmmcpmcp-servermulti-agentollamaopenclaworchestrationself-hosted

Embed the badge

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

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

Questions

Does swarmclaw run?
Yes. swarmclaw runs. Argusic installed and launched it on a clean machine in 43 minutes, hitting 3 errors on the way, and recorded the session.
How did Argusic test swarmclaw?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit ed38ba5329c2. One attempt is recorded, and the full method is on the methodology page.
What did it cost to test swarmclaw?
The run that produced this verdict cost $0.06: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does swarmclaw take to install?
43 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 swarmclaw need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing swarmclaw?
3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
How does swarmclaw compare with the alternatives?
Of the 64 agent-framework projects Argusic has installed and timed, swarmclaw was the 63rd fastest to reach a running state, and 51 of 64 reached one at all.
Where is the evidence for swarmclaw?
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