Argusic daily report, 1 October 2026

On 1 October 2026 (UTC) an AI agent tried to install and run 96 open-source projects on clean machines, with nobody helping: 105 attempts in all, 2 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 2026-10-02.

The day in numbers

Projects: 96
Attempts: 105
Runs: 63
Runs with mocks: 32
Could not verify: 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.

ANUS6,551 stars, runs
dsh-market5,239 stars, runs
devspace5,183 stars, runs
bumblebee5,049 stars, runs
dbhub3,593 stars, could not verify

All 96 projects, most starred first

subject	verdict	topic	stars	attempts	evidence
ANUS	Runs	ai-agent	6,551	1	log and recording
dsh-market	Runs	deepseek-harness	5,239	1	log and recording
devspace	Runs	agent-orchestration	5,183	1	log and recording
Raven	Runs with mocks	ai	5,050	1	log and recording
bumblebee	Runs	golang	5,049	1	log and recording
anti-slop	Runs	agent-skills	5,046	1	log and recording
FrontierAgent	Runs with mocks	agent-orchestration	4,876	1	log and recording
niubigeo	Runs with mocks	TypeScript	4,859	1	log and recording
skills	Runs with mocks	Python	4,803	1	log and recording
animal-island-ui	Runs	component-library	4,728	1	log and recording
SemIf-OpenJev	Runs with mocks	Python	4,648	1	log and recording
mattpocock-skills-zh-CN	Runs	JavaScript	4,561	1	log and recording
skills	Runs	skills	4,508	1	log and recording
esp32-ai	Runs	Python	4,445	1	log and recording
nub	Runs	javascript-runtime	4,364	1	log and recording
ziwei-doushu	Runs	bazi	4,325	1	log and recording
Zappa	Runs with mocks	alb	3,698	2	log and recording
kiali	Runs	istio	3,642	1	log and recording
boost	Runs	ai	3,640	1	log and recording
quiver	Runs	category-theory	3,627	1	log and recording
express-status-monitor	Runs	express	3,615	1	log and recording
oxidized	Runs	backup	3,596	1	log and recording
llm-wiki-agent	Runs	ai-agent	3,595	2	log and recording
dbhub	Could not verify	agents	3,593	2	log and recording
buildwithclaude	Runs with mocks	claude	3,574	1	log and recording
MyPerf4J	Runs with mocks	agent	3,567	1	log and recording
toapi	Runs	api	3,556	1	log and recording
proxy.py	Runs	dns-over-https	3,551	1	log and recording
DoraCMS	Runs with mocks	cms	3,545	1	log and recording
mcp-grafana	Runs	Go	3,521	2	log and recording
cost-xray	Runs	claude-code	3,520	1	log and recording
playwright-go	Runs	automation	3,513	1	log and recording
obsidian-wiki	Runs with mocks	agent-skills	3,512	1	log and recording
wavedrom	Runs	diagram	3,502	1	log and recording
arscontexta	Runs	claude-code	3,493	1	log and recording
py-xiaozhi	Runs with mocks	cross-platform	3,488	1	log and recording
svix-webhooks	Runs with mocks	api	3,424	2	log and recording
tududi	Runs	expressjs	3,406	1	log and recording
rmpc	Runs with mocks	mpd	3,393	1	log and recording
awesome-generative-ai-apps	Runs with mocks	ai-apps	3,357	1	log and recording
whisper-asr-webservice	Runs	asr	3,349	1	log and recording
paths-filter	Runs	change-detection	3,348	1	log and recording
hl	Runs	cli	3,300	1	log and recording
php-curl-class	Runs	api	3,299	1	log and recording
rustic	Runs	backup	3,293	1	log and recording
neo	Runs	agent-memory	3,283	1	log and recording
davinci-resolve-mcp	Runs	blackmagic	3,281	1	log and recording
TTS-WebUI	Runs	ace-step	3,279	1	log and recording
pezzo	Runs with mocks	ai	3,277	1	log and recording
google_workspace_mcp	Runs with mocks	ai	3,273	1	log and recording
fastmcp	Runs	mcp	3,272	1	log and recording
tuicr	Runs	ai-tools	3,267	1	log and recording
setup-php	Runs with mocks	code-coverage	3,265	1	log and recording
graphic-walker	Runs	bi	3,258	1	log and recording
open-terminal	Runs	agentic-ai	3,256	1	log and recording
claude-tap	Runs	agent-debugging	3,254	1	log and recording
squad	Runs	ai	3,249	1	log and recording
dblab	Runs with mocks	cli	3,241	2	log and recording
ruby-openai	Runs with mocks	ai	3,222	1	log and recording
Gladys	Runs	assistant	3,218	1	log and recording
blazingmq	Runs	blazingmq	3,215	2	log and recording
CudaText	Runs	cross-platform	3,211	1	log and recording
cortex	Runs	api-client	3,211	1	log and recording
fli	Runs with mocks	api	3,201	1	log and recording
tntsearch	Runs	full-text-search	3,199	1	log and recording
analog	Runs	angular	3,187	1	log and recording
sh	Runs with mocks	automation	3,187	2	log and recording
arxiv-mcp-server	Runs	ai	3,185	1	log and recording
retina	Runs	ebpf	3,177	1	log and recording
humanize-text	Runs with mocks	ai-humanizer	3,170	1	log and recording
organize	Runs	automatic	3,157	1	log and recording
tilemill	Runs with mocks	mapbox	3,150	1	log and recording
vps-audit	Runs	auditi	3,150	1	log and recording
browser-control	Runs	agent	3,147	1	log and recording
kombu	Runs	celery	3,144	2	log and recording
netbeans	Runs	code-editor	3,141	1	log and recording
gluesql	Runs	database	3,131	1	log and recording
ableton-mcp	Runs with mocks	ableton	3,127	1	log and recording
elevenlabs-python	Runs with mocks	artificial-intelligence	3,125	1	log and recording
vibe-coding-prompt-template	Runs	ai-agents	3,121	1	log and recording
engine	Runs	3d-engine	3,117	1	log and recording
MiroFlow	Runs with mocks	agent-framework	3,116	1	log and recording
termscp	Runs with mocks	aws-s3	3,110	1	log and recording
agent-scan	Runs with mocks	agent	3,107	1	log and recording
GPA	Runs	asr	3,106	1	log and recording
robusta	Runs with mocks	alerting	3,104	1	log and recording
asciigraph	Runs	ascii-chart	3,097	1	log and recording
subtrace	Runs with mocks	backend	3,097	1	log and recording
photon	Runs	elasticsearch	3,088	1	log and recording
dnote	Runs	cli	3,086	1	log and recording
http	Runs	client	3,082	1	log and recording
skills	Runs with mocks	agent-skills	3,070	1	log and recording
gateway	Runs with mocks	api-gateway	3,064	1	log and recording
gobackup	Runs	backup	3,063	1	log and recording
cc-gateway	Runs	anthropic	3,056	1	log and recording
swirl-search	Runs	ai-search	3,047	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.