Argusic daily report, 20 October 2026

On 20 October 2026 (UTC) an AI agent tried to install and run 84 open-source projects on clean machines, with nobody helping: 98 attempts in all.

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 2026-10-21.

The day in numbers

Projects: 84
Attempts: 98
Runs: 53
Runs with mocks: 25
Could not verify: 5
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-browser44,332 stars, runs
genkit6,532 stars, runs
JobSpy4,520 stars, runs
youtube-automation-agent4,489 stars, runs
scrawl570 stars, could not verify

All 84 projects, most starred first

subject	verdict	topic	stars	attempts	evidence
agent-browser	Runs	Rust	44,332	1	log and recording
genkit	Runs	agents	6,532	1	log and recording
JobSpy	Runs	bayt	4,520	1	log and recording
youtube-automation-agent	Runs	ai-agents	4,489	1	log and recording
craft-launcher	Runs	Rust	3,644	1	log and recording
Mirrai	Runs with mocks	ai-chatbot	1,289	1	log and recording
scrawl	Could not verify	ai-agents	570	2	log and recording
nvx	Runs with mocks	agentic-coding	546	1	log and recording
sno-station-skills	Runs with mocks	agent-handoff	510	1	log and recording
jarvis-assistant-vocal	Runs with mocks	anthropic-claude	486	1	log and recording
Flowtrace	Runs	agent-skills	485	1	log and recording
CodeAF	Runs	agent-fleet	484	1	log and recording
flyto-core	Runs with mocks	ai-agents	484	1	log and recording
Connexio	Runs	developer-tools	483	1	log and recording
10xProductivity	Runs with mocks	accountability	482	1	log and recording
HaCasa	Runs with mocks	custom-panel	481	1	log and recording
dbflux	Could not verify	aws	481	2	log and recording
next-ecommerce	Runs	apollo	481	1	log and recording
pigo	Runs with mocks	coding-agent	481	1	log and recording
backup-x	Runs	backup	480	1	log and recording
fundus	Runs	cc-news	480	1	log and recording
laravel-transporter	Runs	api-client	479	1	log and recording
iot-inspector-client	Runs with mocks	internet-of-things	478	1	log and recording
Quantum	Runs	automation	477	1	log and recording
talkcody	Runs with mocks	ai-agents	477	1	log and recording
ViNote	Runs	agent-harness	475	1	log and recording
catbus	Runs with mocks	ai-agent	475	1	log and recording
gwq	Runs	ai	475	1	log and recording
platform-ststm32	Runs	arduino	475	1	log and recording
Async	Runs	ai-coding-assistant	474	1	log and recording
deepseek-php-client	Runs with mocks	ai	473	1	log and recording
WinShun	Could not verify	chinese	472	2	log and recording
aeria	Runs	aeria	472	2	log and recording
azuretls-client	Runs	golang	472	1	log and recording
chief	Runs	ai	472	1	log and recording
cli	Runs with mocks	ai-agent	472	1	log and recording
db2rest	Runs	ai	472	2	log and recording
shopkeeper-agent	Runs with mocks	agent-workflow	472	1	log and recording
wikipedia-semantic-search	Runs with mocks	ai	472	1	log and recording
design-studio	Runs	automation-tool	471	1	log and recording
osm	Runs	golang	470	1	log and recording
rapidgzip	Runs	cli	470	1	log and recording
snip	Runs	ai-coding	470	1	log and recording
NASSAV	Runs with mocks	jable	469	2	log and recording
deepgram-python-sdk	Runs with mocks	asr	469	1	log and recording
koji	Runs	cli	469	1	log and recording
hass-opnsense	Runs with mocks	hass	467	1	log and recording
pyngrok	Runs	localhost	466	1	log and recording
qdrant-js	Runs	javascript	466	1	log and recording
claudeblattman	Runs	academic-research	465	1	log and recording
cms	Runs with mocks	api-gateway	465	1	log and recording
AI-Cross-Border-Ecommerce-Agent-Toolkit	Runs	ai-agents	464	1	log and recording
cortexdb	Runs	embeddings	464	1	log and recording
filament-kanban	Runs	filament-plugin	464	1	log and recording
github-trending-backup	Runs	backup	464	1	log and recording
irrigation_unlimited	Runs	garden-automation	464	1	log and recording
daffodil	Runs	angular	463	1	log and recording
Linghun	Runs with mocks	agent-runtime	462	1	log and recording
free-pmo	Runs	freelancer	462	1	log and recording
home-assistant-global-health-score	Runs	architecture	462	1	log and recording
airAnime	Runs	anime	461	1	log and recording
cersei	Runs	agent	461	2	log and recording
circleci-cli	Runs	circle-ci	461	1	log and recording
gravitee-api-management	Runs	api	461	1	log and recording
tinylittleoffice	Could not verify	cloudflare-workers	461	2	log and recording
lineage-skill	Runs	course-distillation	460	1	log and recording
obsidian-plugin	Runs	notes	460	1	log and recording
Spire	Runs	api-gateway	459	1	log and recording
agentsilex	Runs	agent-framework	458	2	log and recording
bux	Runs with mocks	ai-agent	458	1	log and recording
react-native-vlc-media-player	Could not verify	audio	458	2	log and recording
html2image	Runs	chrome	455	1	log and recording
agents	Runs with mocks	agentic-workflow	454	1	log and recording
diri	Runs	claude-code	454	1	log and recording
2fa	Runs with mocks	2fa	453	1	log and recording
CharDesk	Not yet verified	ai-context	453	1	log and recording
ProjectAtlas	Runs	claude-code	453	1	log and recording
VRCT	Runs	osc	453	2	log and recording
percho	Runs	agent	453	2	log and recording
python-sdk	Runs with mocks	pinecone	453	2	log and recording
AutoHarness	Runs	agent-framework	452	1	log and recording
action	Runs	actions	450	1	log and recording
beever-atlas	Runs with mocks	agent-memory	450	2	log and recording
miniblue	Runs	azure	450	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.