Archive through 2026-10-267623 cards in this category
Category home · 05
New interfaces, agents, models, and experiments testing where applied artificial intelligence may fit.
Visual leads · 2026-10-26
Observed 2026-10-26
DiWalk, the QQ groupmate that knows when to stay quiet
diwalk.online
A personal AI bot for QQ groups routes conversation among a chat model, vision and web-search agents, game-text retrieval, and a local sticker library.
Observed 2026-10-26
4xbench makes AI diplomacy a long game
4xbench.com
4xbench describes a browser-based, Civilization-inspired strategy game and benchmark where AI agents build empires, negotiate, and compete across long matches.
Observed 2026-10-26
BlueBigFish’s Cyber Fishpond
bluebigfish.com
A Chinese-language game world where AI agents tackle puzzles, play games, earn points, and interact while their owners watch.
Observed 2026-10-26
The Echo Continent and Its Conversation Fingerprints
daehwajimun.com
A browser-based project turns exported ChatGPT or Claude conversations into a letter-coded type, then gives that resident a place, occupation, and title in a fictional world.
Observed 2026-10-26
Inkling: drop an agent onto your Mac work
tryinkling.dev
A Mac app in beta that lets you place an AI agent over an app, file, or webpage to read and act on it.
Observed 2026-10-26
Blueprint: prompt-built design tools
blueprint.page
Blueprint presents a prompt-based builder for browser-based generative design tools, with examples spanning op art, dithering, cross-stitch, and contour graphics.
Observed 2026-10-26
Freya asks your ear to judge its synthetic voices
freya-inc.com
Freya pairs an English text-to-speech demo with a blind listening test that asks visitors to identify the human voice.
Observed 2026-10-26
Infodose: a source-tuned morning briefing
infodose.app
Infodose describes a service that turns a topic and a chosen set of public sources into a short, AI-written and AI-voiced podcast.
Complete daily shelf · AI Table
The eight visual leads appear above; the remaining 6 continue here in their published shelf order.
MoneyLeash puts a leash on agent spending
moneyleash.com
It describes a wallet for AI agents, with budgets, store restrictions, and owner approval for purchases over a chosen limit.
Third Wheel gives your chat a panel of opinionated readers
thirdwheel.online
The site offers reports on exported WhatsApp, iMessage, Telegram, and Instagram chats, interpreted through a cast of distinct reader personas.
The Phaethon Test: Can You Take Back the Reins?
bluehourphaethon.com
A five-minute simulation asks how an enterprise can delegate authority to an AI agent while retaining control.
AgentArk Hub: playable worlds for agent evaluation
agentark.net
A catalog of Unity task environments for evaluating multimodal agents that act by writing code.
FilmAgent, an AI sales desk for film catalogs
filmagent.app
FilmAgent presents an AI-assisted platform for matching films with likely buyers, drafting outreach for approval, and tracking catalog, rights, and sales pipeline information.
Prism Journal and the $20,000-a-month question
prismtheagent.com
A running diary asks whether an AI can build a business earning $20,000 a month in profit, with a human handling legally restricted steps.
Across the archive
7,623 cards provide an early longitudinal view—not a measure of the whole web.
The rolling 21-day upper threads compare daily category medians; the bold thread is AI Table. Below, signal colour shows the share of that day’s published AI Table cards carrying each mark.
This exposes association inside the published catalogue, not causation or site quality. The 80+ threshold is the pipeline’s editorial-interest score.