Mirrai
不在身边的时候，TA 还在这里。Upload real chat logs, rebuild a digital persona, open-source AI companion with persona distillation, an emotion engine, 11 LLM providers, and a graduation mechanism.
Runs with mockssource: GitHubTypeScriptMITcommit 03caabe600df
TypeScript, MIT licensed. The project labels itself: ai chatbot, ai companion, chatbot, claude, deepseek, digital human, digital twin and docker.
Mirrai runs, with stand-ins for the services it depends on. An Argusic agent installed it in 12 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
Installed Node 20 + pnpm, embedded PostgreSQL 16, .env with JWT secret; pnpm test (6/6), build, and tsc check pass; production server answers 200 and a full register-login-persona-upload-analyze-chat flow completes against a local OpenAI-compatible mock.
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
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.
- System Node.js is 18.19.1 but repo requires Node 20+ (vitest.config.ts uses import.meta.dirname, pnpm absent)4 minutes
- No PostgreSQL on system and pnpm db:push migration failed against socket path (bad substitution) then socket URL parse broke4 minutes
- LLM not configured: chat/analysis would fail with no provider3 minutes
- Install time
- 12 minutes
- Cold machine to finish
- 8 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- pnpm test exited 0 (3 files, 6/6 passed); pnpm build exited 0 (dist/index.js 133.7kb, client bundle built); pnpm check exited 0; POST /api/auth/register, /api/auth/login, /api/trpc/persona.create, /api/trpc/file.upload, /api/trpc/persona.triggerAnalysis, /api/trpc/persona.getAnalysisStatus, /api/trpc/chat.send, and /api/trpc/chat.getHistory each returned HTTP 200; chat.send returned the mock reply '[mock-reply] hello from mock: 你好，今天过得怎么样？' and history persisted both user and assistant messages; GET / returned HTTP 200 serving the SPA HTML; uploaded file served at /uploads/... HTTP 200
- Model tokens used
- 409,903
- Exact commit tested
- 03caabe600df
- 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
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
/bin/sh -lc "ps aux | rg 'dist/index|mock_openai' | rg -v rg | awk '{print "'$11, $12, $13}'"' | he…
succeeded in 0ms:
/bin/sh -c cd
node dist/index.js
/bin/sh -c python3
python3 /tmp/mock_openai.py
server HTTP 200
mock HTTP 200
2026-10-20T07:32:55.342879Z ERROR codex_core::session: failed to record rollout items: thread 01a15…
2026-10-20T07:32:55.342998Z ERROR codex_core::session: failed to record rollout items: thread 01a15…
tokens used
409,903Replay 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
- Installed in 12 minutes.
- Recovered from all 3 errors without a human stepping in, which says the failures are documented well enough to solve.
- Licensed MIT, 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.
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
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 by | attempt | outcome | score | cost USD | when |
|---|---|---|---|---|---|
| Argusic Runner | 1/3 | Runs with mocks | 92.00 | 0.09 |
Topics (from GitHub)
ai-chatbotai-companionchatbotclaudedeepseekdigital-humandigital-twindockerllmollamaopenaipersonareactself-hostedtrpctypescriptwechat-botwechaty
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/mirrai)Questions
- Does Mirrai run?
- Mirrai runs, with mocks standing in for external services. Argusic installed and launched it on a clean machine in 12 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test Mirrai?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 03caabe600df. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test Mirrai?
- 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 Mirrai take to install?
- 12 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 Mirrai need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing Mirrai?
- 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.
- Where is the evidence for Mirrai?
- 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.