Argusic daily report, 19 October 2026

On 19 October 2026 (UTC) an AI agent tried to install and run 85 open-source projects on clean machines, with nobody helping: 121 attempts in all, 3 of them cut short by our own tooling and not counted.

Verdicts as they stood at the end of that day. A project tested again since may have moved; its own page always shows the current verdict. Written 2026-10-20.

The day in numbers

Projects: 85
Attempts: 121
Runs: 42
Runs with mocks: 20
Could not verify: 22
Not yet verified: 1

Five from the day

Chosen by a fixed rule, not by hand: the three most starred projects of the day, then the most starred one that runs, then the most starred one that could not be verified.

agent-orchestrator13,367 stars, could not verify
harness210,955 stars, could not verify
replica-skill3,517 stars, runs
suneditor2,021 stars, runs
codefever2,750 stars, could not verify

All 85 projects, most starred first

subject	verdict	topic	stars	attempts	evidence
agent-orchestrator	Could not verify	agent-fleet	13,367	2	log and recording
harness2	Could not verify	Python	10,955	2	log and recording
replica-skill	Runs	agent	3,517	1	log and recording
codefever	Could not verify	docker	2,750	2	log and recording
suneditor	Runs	cms	2,021	2	log and recording
infinite-image-browsing	Runs	audio	1,359	1	log and recording
blender-ai-mcp-addon	Runs	ai-assistant	930	1	log and recording
agent-browser-cli	Runs with mocks	agent	635	1	log and recording
claude-remotion-skill	Runs	ai-agents	541	1	log and recording
Rythe-Engine	Could not verify	3d-audio	534	1	log and recording
bug-hunter	Could not verify	adversarial-ai	526	2	log and recording
homeassistant-solax-modbus	Runs	alphaess	526	1	log and recording
Extract	Could not verify	archive	525	2	log and recording
gitlabber	Runs with mocks	backup	525	1	log and recording
acebase	Runs	acebase	524	1	log and recording
fable51-worlds	Runs	3d-reconstruction	523	1	log and recording
ha-solarman	Could not verify	afore	523	2	log and recording
unbody	Could not verify	agentic-ai	523	2	log and recording
AgenticX	Could not verify	agent-framework	521	2	log and recording
lychee-action	Runs	action	521	1	log and recording
django-distill	Runs	django	519	2	log and recording
nextplate	Could not verify	boilerplate-template	519	2	log and recording
performance	Could not verify	analyzer	519	2	log and recording
windows-capture	Could not verify	game-capture	519	2	log and recording
kition	Could not verify	agent	518	2	log and recording
nylon	Runs	distributed-networking	518	2	log and recording
ComfyUI-VoxCPM	Runs with mocks	ai-voice	517	2	log and recording
project-management	Runs	full-stack	517	1	log and recording
roam-code	Could not verify	ai-agents	517	2	log and recording
CatchMe	Could not verify	ai-agent	516	2	log and recording
models	Runs	ai	516	2	log and recording
qbittorrent-api	Runs with mocks	api	516	1	log and recording
soundcli	Not yet verified	cli	516	3	log and recording
gronx	Could not verify	adhocore	515	2	log and recording
ESPAltherma	Runs	arduino	513	1	log and recording
mcp-ssh-manager	Runs with mocks	anthropic	512	1	log and recording
octocode	Runs	ai	512	1	log and recording
pi-pocket	Runs	ai-agent	512	1	log and recording
bramble	Runs	api-gateway	511	1	log and recording
machinist	Could not verify	ai-agents	511	2	log and recording
maestro-runner	Could not verify	android	511	2	log and recording
whirl	Runs with mocks	ai	511	1	log and recording
libvlc-go	Runs	audio	510	1	log and recording
nats.c	Runs	c	510	1	log and recording
orca-agent	Could not verify	ai-agents	510	2	log and recording
routeros-api-php	Runs with mocks	api-client	508	1	log and recording
panel	Runs with mocks	billing	507	2	log and recording
Audar-ASR-V1	Runs	arabic	506	1	log and recording
MusicBox	Runs	electron	506	1	log and recording
static-curl	Runs	curl	506	2	log and recording
scrapple	Runs with mocks	beautifulsoup	505	1	log and recording
ReplicaDB	Runs with mocks	cross-platform	504	1	log and recording
heh	Runs	command-line	504	1	log and recording
python-calamine	Runs	excel	504	1	log and recording
ipc	Runs	async	503	1	log and recording
watch-skill	Could not verify	agent-observability	503	2	log and recording
flock	Runs	ai-agent	502	1	log and recording
vibe-check-mcp-server	Runs	agentic-ai	502	1	log and recording
Ghost-CLI	Runs	blogging	499	1	log and recording
meridian	Runs	3x-ui	499	1	log and recording
opencrew	Runs	a2a-protocol	499	1	log and recording
firestore-backup-restore	Runs with mocks	backup	498	1	log and recording
aspeak	Runs with mocks	aspeak	497	1	log and recording
ytdlp2STRM	Runs	crunchyroll	497	2	log and recording
qmarkdowntextedit	Could not verify	cpp	496	2	log and recording
opencode-goal-plugin	Runs	agent	495	2	log and recording
s3-proxy	Runs with mocks	basic-authentication	495	1	log and recording
innodb-java-reader	Runs	command-line-tool	494	1	log and recording
mirrormate	Runs with mocks	ai	494	1	log and recording
docker-vackup	Runs with mocks	backup	493	2	log and recording
grok-keysmith	Runs with mocks	agents-md	493	1	log and recording
mantrae	Runs	go	493	1	log and recording
lnxlink	Runs with mocks	automation	492	1	log and recording
CoWork-OS	Runs	ai-assistants	491	2	log and recording
better-design	Runs	ai-design	491	1	log and recording
Kiri	Runs	cli	490	1	log and recording
Weibo-archiver	Could not verify	archiver	489	2	log and recording
hyde	Runs	blog-engine	489	1	log and recording
jietuba	Runs with mocks	clipboard-history	489	2	log and recording
umya-spreadsheet	Runs	excel	489	1	log and recording
discord-backup	Runs with mocks	backup	487	2	log and recording
ha-evcc	Runs with mocks	evcc	487	1	log and recording
FutureSDR	Runs	dsp	486	1	log and recording
core	Runs	ai	486	1	log and recording
reason-network	Runs	coding-agents	485	1	log and recording

How a verdict is reached: the methodology. Every project tested so far: the subjects. GPU cloud and VPS prices: servers. Laptops and desktops for programming and AI: hardware. Earlier days: the daily archive.