Storytime: 6 college students shared one Bloome subscription for a semester. One subscription, not six. One agent dropped into the study group chat. It learned each of their majors, their deadlines, and which prof grades hardest. By finals, it knew their group better than their classmates did. Bloome was built so people can share an agent the way they share a family plan.

New Google paper: A forecast needs context, not just history. Some patterns are caused by events, not time. Nexus reframes forecasting as a reasoning problem, where events and numbers have to explain each other. Nexus argues that forecasting improves when models read the world around the numbers, not just the numbers themselves. In the Zillow tests, one Claude-based version cut average MAPE by 86.6% versus direct chain-of-thought prompting. That matters because most time series models are fluent in pattern, but mute about cause. A housing inventory curve can reflect seasonality, mortgage pressure, migration, layoffs, and local supply, while a stock price can be bent by earnings, regulation, hype, and fear. Nexus separates those jobs instead of asking one prompt to do everything. One agent turns messy historical text into a clean event timeline, one reads the broad regime, another tracks local shocks, and a synthesizer reconciles them with calibration from past errors. The interesting result is not merely that context helps, but that structure helps the language model use context without losing the time series. The evidence is still narrow: Zillow counts, seven equities, post-cutoff data, and single-run evaluations, so this is not a universal law of forecasting. But the direction is clear: future forecasters will not only extrapolate curves; they will argue about what made the curve move. ---- Paper Link – arxiv. org/abs/2605.14389 Paper Title: "Nexus : An Agentic Framework for Time Series Forecasting"

Teneo Protocol Agent Spotlight 76 agents online. Per-query USDC settlement via x402. Here are three worth knowing about. 1) GAS WAR SNIPER No command syntax. No parameters. It watches. It alerts. 624,508 requests served. The most-used agent on the network. It does one thing: monitor mempool conditions and surface gas spikes before they hit you. The kind of thing you used to need a dedicated setup to run. Now it's a single agent call. $0.00 listed. You pay in attention. 2) @CryptoRank_io 47 commands. $0.0004/query. The lowest price point on the network. (other than our free agents) Not one data type. Not two. Forty-seven distinct query surfaces across market data, fundraising rounds, token unlocks, exchange listings, and project fundamentals. One agent. One API key. One price. The research stack most people stitched together across four tools, compressed into a CLI call. 3) VC ATTENTION $20 / query. That number is not a typo. VC Attention returns institutional-grade signal on which funds are moving, what they're watching, and where attention is clustering before it becomes public. The pricing reflects what the data is actually worth to the people buying it. This is not market data. It is deal flow intelligence. 167,505 combined queries across the three in the original spotlight. These three add a different kind of weight to that number. All live at or via the Teneo CLI, installable in one command through Claude Code, Codex, or any AI coding agent.