Companion-Space

本地优先的二次元陪伴学习应用 · Local-first anime companion and study app.

Runssource: GitHubPythonApache-2.0commit 52543119e642

Python, Apache-2.0 licensed. The project labels itself: ai companion, fastapi, local first, nextjs, rag, self hosted, voice ai and vrm.

Companion-Space runs. An Argusic agent installed it in 8 minutes and hit no errors on a clean machine with no GPU, and the whole session was recorded.

The backend API starts, all 549 Python tests pass, the Next.js frontend typechecks, lints, and builds, and the mock end-to-end flow (vault unlock, space creation, auto-assigned mock connections, text and streaming turn with valid structured responses, session end) works correctly without any real API keys.

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
$0.03 (measured)
recorded runs
2
last tested
stars
363
forks
0
open issues
0
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.

Install time
8 minutes
Cold machine to finish
7 minutes
Errors hit and fixed
0 hit, 0 fixed with no human help
How the result was proved
pytest services/api/tests -q returned 549 passed, 0 failed. curl -s -o /dev/null -w '%{http_code}' http://127.0.0.1:8000/healthz returned 200. Full mock end-to-end flow verified: vault init/unlock (200), space creation (201), provider registry (200 with 7 providers), auto-assigned mock connections (4 assignments created on space bootstrap), text turn via POST /sessions/{id}/turns (200, valid mock reply with citations=[], emotion, board_actions), streaming turn via POST /sessions/{id}/turns/stream (200, SSE-style llm.delta events), session end (200). npm run build:web compiled all 16 routes successfully. npm run typecheck:web (tsc --noEmit) passed. npm run lint:web (eslint --max-warnings=0) passed.
Model tokens used
135,894
Exact commit tested
52543119e642
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.

/bin/sh -lc 'cd /work/repo && export PATH="/tmp/node-v24.0.0-linux-x64/bin:$PATH" && node scripts/c…
 succeeded in 481ms:
Pet asset integrity passed: mori, yuzu
Let me compile the final report. Here's a summary of what was done and observed.
exec
/bin/sh -lc '# Clean up any remaining server
pkill -f uvicorn 2>/dev/null; true' in /work/repo
 exited 143 in 0ms:
All verification passes. Here's my report:
tokens used
135,894
All verification passes. Here's my 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.
  • Installed in 8 minutes.
  • Nothing broke on the way: zero errors between clone and running.
  • Ran without a GPU, so it does not need one to start.
  • Licensed Apache-2.0, as reported by its host.
  • Measured 2 times, so the result is not a one-off.

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 8 minutes after the clone.
  • Following its own README: the documented path worked with nothing to fix.

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 Runner2/3Runs100.000.03
Argusic Runner1/3Runs100.000.03

Topics (from GitHub)

ai-companionfastapilocal-firstnextjsragself-hostedvoice-aivrm

Embed the badge

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

[![Tested by Argusic](https://argusic.com/badge/Companion-Space.svg)](https://argusic.com/subject/companion-space)

Questions

Does Companion-Space run?
Yes. Companion-Space runs. Argusic installed and launched it on a clean machine in 8 minutes, hitting 0 errors on the way, and recorded the session.
How did Argusic test Companion-Space?
On a fresh, disposable machine, with every command recorded and the repository pinned at commit 52543119e642. 2 attempts are recorded, and the full method is on the methodology page.
What did it cost to test Companion-Space?
The run that produced this verdict cost $0.03: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
How long does Companion-Space take to install?
8 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 Companion-Space need a GPU?
Not to start. Argusic reached a running state on a machine with no GPU.
What goes wrong when installing Companion-Space?
Nothing did, in the recorded run: zero errors between clone and running.
Where is the evidence for Companion-Space?
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.

Discussion