Argusic daily report, 15 October 2026

On 15 October 2026 (UTC) an AI agent tried to install and run 71 open-source projects on clean machines, with nobody helping: 81 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 2026-10-16.

The day in numbers

Projects: 71
Attempts: 81
Runs: 48
Runs with mocks: 18
Could not verify: 2
Not yet verified: 3

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.

zod44,135 stars, runs
quotio4,892 stars, runs with mocks
codex-plusplus3,754 stars, could not verify
agent-memory3,753 stars, runs
docker-mac-net-connect666 stars, could not verify

All 71 projects, most starred first

subject	verdict	topic	stars	attempts	evidence
zod	Runs	runtime-validation	44,135	2	log and recording
quotio	Runs with mocks	ai-tools	4,892	1	log and recording
codex-plusplus	Could not verify	JavaScript	3,754	2	log and recording
agent-memory	Runs	agent-memory	3,753	1	log and recording
openmeld	Runs	agent-client-protocol	1,359	1	log and recording
forkcms	Runs	cms	1,172	1	log and recording
awesome-agent-apis	Runs with mocks	agent-apis	1,055	1	log and recording
scout	Runs	ai-agent	830	1	log and recording
nuki_hub	Not yet verified	bluetooth	747	1	log and recording
codeAgent	Runs	agent	741	1	log and recording
screen_capture_lite	Runs	c-plus-plus	727	1	log and recording
MegaMemory	Runs	agentic-coding	725	1	log and recording
opik-openclaw	Runs	ai-agents	725	1	log and recording
node-sonos	Runs with mocks	home-automation	724	1	log and recording
claude-chat-exporter	Runs with mocks	ai-chat	723	1	log and recording
otel-cli	Runs	cli	723	1	log and recording
jafka	Runs	jafka-mq	720	1	log and recording
reactpress	Runs	cms	720	1	log and recording
scrapecraft	Runs	ai	719	1	log and recording
MCP-Nest	Runs	llm	718	1	log and recording
mcp-server-elasticsearch	Runs with mocks	elasticsearch	718	1	log and recording
openqodex	Runs with mocks	agent-skills	715	2	log and recording
Adobe_Premiere_Pro_MCP	Runs with mocks	adobe	702	1	log and recording
codemap	Runs	claude	702	1	log and recording
awl	Runs	cross-platform	700	1	log and recording
ytermusic	Runs	music-player	699	2	log and recording
EcommerceAPI	Runs with mocks	backend	698	1	log and recording
algoliasearch-client-php	Runs with mocks	algolia	696	1	log and recording
tplay	Runs	ascii	696	1	log and recording
obsidian-mcp-server	Runs with mocks	ai-tools	695	1	log and recording
expo-speech-recognition	Runs with mocks	expo	693	1	log and recording
pi-extensions	Runs	ai-agents	693	1	log and recording
react-h5-audio-player	Runs	audio	693	1	log and recording
deepcrawl	Runs	ai-agent-tools	690	1	log and recording
openinframap	Not yet verified	gis	689	1	log and recording
swarmclaw	Runs	agent-framework	689	1	log and recording
agent-safe-pipeline	Runs	agent-security	688	1	log and recording
azure-devops-cli-extension	Runs with mocks	agile	688	2	log and recording
cmd2	Runs	auto-completion	688	1	log and recording
voxelize	Runs	baltaazr	688	1	log and recording
ApplicationInsights-JS	Runs	analytics	687	1	log and recording
Prism-Browser-Community	Runs	browser-automation	687	1	log and recording
aislop	Runs	ai	687	1	log and recording
ghapi	Runs	api-client	687	1	log and recording
klog	Runs	cli	686	1	log and recording
php-censor	Runs	ci	686	1	log and recording
comics-downloader	Runs	cbr	685	1	log and recording
shotgun	Runs	cli	685	1	log and recording
comet	Runs	api	684	1	log and recording
Mimi	Not yet verified	android	683	1	log and recording
mimi	Runs with mocks	audio-capture	683	1	log and recording
openlrc	Runs with mocks	auto-subtitle	681	1	log and recording
Deuz-SDK	Runs	agent-framework	680	2	log and recording
agentos	Runs	agent-framework	678	1	log and recording
jsongrep	Runs	cbor	677	1	log and recording
mcp-client-cli	Runs with mocks	langchain	675	1	log and recording
paozhu	Runs	http-client	673	1	log and recording
vmprint	Runs	bidirectional-text	673	1	log and recording
RAGLight	Runs	agentic-ai	672	1	log and recording
jellyfin-plugin-skin-manager	Runs	jellyfin	671	3	log and recording
better-deepseek	Runs with mocks	ai-tools	670	1	log and recording
second-brain	Runs	agentic-ai	670	1	log and recording
ui-builder	Runs	no-code	670	1	log and recording
ngx-scanner	Runs	angular	668	1	log and recording
tunnl.gg	Runs	expose	668	1	log and recording
ai-development-patterns	Runs	ai-assisted-coding	666	1	log and recording
chart-releaser-action	Runs with mocks	actions	666	1	log and recording
docker-mac-net-connect	Could not verify	colima	666	2	log and recording
restic-automatic-backup-scheduler	Runs	backup	664	1	log and recording
iwf	Runs with mocks	distributed-process	663	1	log and recording
spring-ai-agent-utils	Runs	agentic-patterns	662	2	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.