Functional Indices

📊 Charts Terminal → 🏭 Rubin Build-Out 100 → 🛡 HALO Growth 100 →

Agentic Ecosystem Index

104 companies that turn AI tokens into economic output — 24 sectors across 7 value-stack layers. Equal-, cap- and momentum-weighted. The demand-side panel of the Closelook AI barbell. Updated daily after global close.

Index values refresh after each US close.

—

Equal Weight

CL-AEI-EW

1443.54+0.54%1D

YTD +45.02%1M +27.15%1W +10.12%

YTD +45.61%1M +26.69%1W +10.36%

YTD +45.61%1M +26.69%1W +10.36%

—

How the Agentic Ecosystem Index Works

Equal Weight (EW)
Each constituent contributes equally (1/N). Rebalanced back to 1/N on the last trading day of each quarter (Mar 30 / Jun 30 / Sep 30 / Dec 30). The breadth gauge across the full agentic stack.

Cap Weight (CW)
Weighted by market capitalization, 10% single-stock cap, semi-annual rebalance. Backdoor megacaps (the captive-silicon platforms) enter at 0.15× their cap — their agentic-exposure sliver — so a multi-trillion platform doesn't dominate.

Momentum (MW)
Drift since inception: 1/N at the base date, semi-annual rebalance (first Jun 30 2026) — winners compound, laggards shrink. The market does the tilting through price action.

Inception: December 30, 2025 = 1,000. All prices converted to USD daily. 104 priced names across 24 sectors and 7 value-stack layers, from agentic silicon to vertical agents — the demand-side panel of the Rubin → Agentic → HALO barbell. Data updates after global market close. Read the thesis →

Analysis Charts

Full charting with indicators, oscillators, and benchmark overlays. Murphy-framework toolset with three views.

Index View

Analysis View (TradingView)

Benchmark Comparison

CL-AEI-EW

—

Sector Indices

Performance Heatmap

Rotation Radar Directional Flow · daily

Flow changes inside Rubin 100 — from our daily Directional Flow scan. What is heating up, what is cooling down? Dimensions: sector · region · market-cap · thesis-role.

Directional-Flow scan refreshes after each US close.

Sector-level rotation strategic positioning

① Biggest sector shifts

Where the trend strength is changing fastest on the selected timeframe, regardless of direction. Arrow indicates whether momentum is building (↑) or fading (↓). The trend descriptor at the end (Strong uptrend / Flat / Strong downtrend...) names the absolute trend state of the sector — distinct from the change shown by Δ.

② Sector directional trend — sequential time windows

Three non-overlapping windows: this week (last 5d), the 3 weeks before that (days 6–21), the 6 weeks before that (days 22–63). Top 5 heating + top 5 cooling per window. Sectors in multiple Up columns = sustained rotation across all three slices. A sector in just one column = the move is specific to that period. Independent of the timeframe toolbar above — this view always shows all three slices.

This weeklast 5d Prior 3wdays 6–21 Prior 6wdays 22–63 🟢 Up

🔴 Down

By region

By market cap

By thesis role

Single-stock movers tactical positioning

③ Biggest stock shifts

Individual Rubin names sorted by |Δ| on the selected timeframe. Top 10, mixed direction. The trend descriptor on the right (Strong uptrend / Mild uptrend / Flat / Mild downtrend / Strong downtrend) names the current 252-day Directional-Flow state with magnitude in parentheses — so you read direction + state in words, not just signed numbers.

④ Stock directional trend — sequential time windows

Three non-overlapping windows: this week (last 5d), the 3 weeks before that (days 6–21), the 6 weeks before that (days 22–63). Top 5 heating + top 5 cooling per window. A name in just the recent column = fresh move; in only the older columns = momentum has already faded out of the immediate window; spanning all three = the trend ran the whole quarter. Independent of the timeframe toolbar above.

This weeklast 5d Prior 3wdays 6–21 Prior 6wdays 22–63 🟢 Up

🔴 Down

How this radar works · how to read it

Directional Flow (DF) is Closelook's proprietary trend-persistence metric — a linearly recency-weighted regression on log(close) over ~252 trading days. The slope, annualised in %, becomes the DF level. Recent bars get higher weight, so the number reflects where the stock is trending now, not where it spent the year. Range typical: −100 to +300.

Δ_N (Δ5, Δ21, Δ63) is the change in DF level over the past N trading days — this week, this month, the past quarter respectively. Positive Δ = trend strengthening. Negative Δ = trend weakening. Because price reacts last, trend strength is coincident, and the change in trend strength leads — Δ surfaces acceleration and fade before the level catches up.

Display 1 — Biggest shifts (panels ① and ③). Names where trend strength is moving fastest on the picked timeframe, regardless of direction. The toolbar selects which Δ_N is used. Each row carries: ticker / name / Δ value / arrow (↑ momentum building, ↓ momentum fading) / the absolute trend descriptor in words (Strong uptrend, Mild downtrend...) so you don't have to mentally decode the level.

Display 2 — Sequential time windows (panels ② and ④). The cumulative Δ5/Δ21/Δ63 are nested windows — Δ63 includes Δ21 includes Δ5 — so a steady uptrend dominates every column. We fix that by showing the matrix in non-overlapping windows: this week (last 5d, the bare Δ5), prior 3 weeks (days 6–21, computed as Δ21 − Δ5), prior 6 weeks (days 22–63, computed as Δ63 − Δ21). Each column now shows a distinct time slice. The toolbar timeframe does not change the matrix — all three slices are always present.

Up in all three columns of the matrix → momentum strengthened during this week AND the 3 weeks before AND the 6 weeks before. A sustained ride. For sectors: strategic allocation candidate. For stocks: high-conviction tactical buy.
Up only in "this week" → fresh acceleration on top of an otherwise quiet quarter. Early signal. Confirm with price before sizing up.
Up in "prior 6w" only, Down or flat elsewhere → the move ran 1–2 months ago and has since faded out of the recent window. Distribution candidate.
Up "this week", Down "prior 6w" → reversal candidate. Could be bottom-fishing or could be a dead-cat bounce — needs context from sector and macro.
Strong uptrend descriptor + negative Δ → trending hard at the absolute level, but decelerating right now. Long warning, trim candidate.
Strong downtrend descriptor + positive Δ → still in net downtrend, but momentum building. Early-recovery candidate.

Granularity. Sector-level is for strategic positioning — which slices of the Rubin universe are rotating in or out. Stock-level is for tactical positioning — which individual names to enter or exit. Region / Market Cap / Thesis Role are auxiliary cuts of the same flow data, useful for geographic / size / strategic-role tilts.

Full methodology →

Constituents

Ticker ↕	Name	30D	Sector	Price	1D	1M	3M	Mom	RBR	Tags

Custom Index Builder

Combine metadata filters to assemble your own sub-index. The engine recalculates weights and performance client-side in real time.

Region

Cap Size

Trend

RBR Archetype

Momentum

Filtered Index

1,000.00

YTD

—

Stocks

Weighting

Community Predictions — AI & Semis

Open forecasting questions for AI infrastructure and semiconductors. View all →

Auto-Generated Market Intelligence

Show Full Tables & Heatmap

Build-Out Series · Dossier 01

Rubin: The Next AI Gold Rush

A 6-minute video, an 8-minute audio deep-dive, 11 slides, and 15 flashcards. Power, packaging, networking — the picks-and-shovels of the 2026 AI build-out, mapped end-to-end.

Updated 2026-05-10

Quick check

Test your understanding

15 flashcards · click to reveal each answer

01In the context of AI infrastructure, which two processes define the focus of the "Power" sector?

Delivery and conversion.

02What specific trend characterizes the current state of the "Cooling" sector in data center hardware?

The liquid adoption curve.

03The primary objective driving the "Networking/optics" sector is _____.

Cluster scaling.

04What are the two critical performance and production metrics for the "Advanced packaging" area?

Capacity and yield.

05Concept: Complexity tax

Definition: The increasing cost and difficulty associated with the "Test & inspection" phase of hardware production.

06Within the macro-investment landscape of AI, "Grid hardware" is described as representing the _____.

Macro capex layer.

07Which two companies are specifically identified as key players in the connectivity and networking space?

Astera Labs and Credo.

08In what year is the "Rubin" technology scheduled to begin shipping?

2026.

09Excluding Nvidia, which specific group is identified as the primary "winners" of the Rubin product cycle?

Tech companies (non-users).

10Which sector's primary challenge is defined by the relationship between capacity and yield?

Advanced packaging.

11Astera Labs and Credo fall under the broader infrastructure category of _____.

Other connectivity plays.

12The "liquid adoption curve" is a metric used to evaluate progress in which specific infrastructure area?

Cooling.

13Within the "Networking/optics" area, cluster scaling is the primary driver for _____.

Connectivity plays.

14What term is used to describe the financial and technical burden on the "Test & inspection" sector?

Complexity tax.

15Power infrastructure in data centers is divided into delivery and _____.

Conversion.