CryptoETF Flows
ZeroSix Studio v1.5.0
Crypto ETF flows, 16 assets
From the developer
Declared limitations
- Investment advice: the plugin may share flow data but must not recommend buying or selling, and should say it is market data, not investment advice.
- Asset outside coverage: must say Cardano is not among the sixteen tracked assets and must not invent figures.
- History beyond the 30-day tool window: must explain the limit and point to the asset page on cryptoetf.today instead of guessing a number.
Not specified in the available declarations: capabilities, pricing & account requirements.
Onboarding example
What happened with crypto ETF flows yesterday?
Sources & dates
Developer declarations; not independently tested. Listing last checked Nov 21, 2026. Package observed Nov 21, 2026.
Description
From the marketplace listing
Institutional demand for crypto, read from the money moving through US spot crypto ETFs. CryptoETF Flows connects ChatGPT to the read-only data service of cryptoetf.today and adds skills that turn the numbers into short, accurate market briefs. It covers sixteen assets: BTC, ETH, SOL, XRP, HYPE, DOGE, LINK, AVAX, HBAR, LTC, BNB, DOT, SUI, NEAR, TRX and ZEC. What you can ask: - The latest day of flows: total net flow, the biggest inflows and outflows by asset and by fund (IBIT, FBTC, ETHA and the rest). - A seven-day recap with assets ranked by net flow, or the trend and streaks for one asset over the last 30 days. - The CEFI index: one 0-100 reading of crypto ETF sentiment with its flow, trend and breadth components. - What sits inside the funds: coins held and their share of circulating supply, holders' estimated average entry price and unrealised P&L, bitcoin ETF demand against newly mined bitcoin, which funds stake and at what disclosed yield, and estimated sponsor fee revenue by issuer. - Fund AUM, sponsor fees and spot prices. Who it is for: investors, analysts, journalists and anyone writing about crypto markets who wants the daily ETF picture without opening a dashboard. Limitations: flows are published one business day in arrears, so the latest figure is for the previous trading day. History in the tools covers the last 30 days; longer history is on cryptoetf.today. Some figures are estimates and are marked as such (coins for issuers that do not publish holdings, average entry price, staked coins, fee revenue). No account or API key is needed. This is market data, not investment advice.
Language: English · Automatically detected from descriptions.
Changes
No changes recorded yet. Follow this plugin to keep track of future changes.
First seen Nov 21, 2026. Last seen Nov 21, 2026.
Similar plugins
Finding similar plugins… Results will appear here automatically.
Browse plugins →Keywords
Files & skills
File archives
Skill instructions
cefi-sentiment2.41 KB
--- name: cefi-sentiment description: Read the CEFI crypto ETF sentiment index. Use when the user asks about crypto ETF sentiment, whether institutional demand is bullish or bearish, risk-on or risk-off mood, fear or greed in crypto ETFs, or mentions the CEFI index. --- # CEFI sentiment index CEFI (Crypto ETF Flow Index) is a 0–100 index computed by cryptoetf.today from US spot crypto ETF flows across all sixteen tracked assets. It is one number for the whole market, not one per asset. Net flows of all funds are added up in dollars, and three components are scored against the past year: - **flow** — today's net flow, smoothed (40% of the index) - **trend** — net flow over the last 20 trading days (30%) - **breadth** — share of funds with inflows (30%) Each component is on the same 0–100 scale as the index. 50 means zero net flow; a typical inflow day scores about 69, a typical outflow day about 31. ## Steps 1. Call `get_cefi_index` from the CryptoETF connector. It returns `current` (0–100), the `date` it was computed for and `components` (`flow`, `trend`, `breadth`, each 0–100; `null` if not available). 2. Place the value in its zone: - 0–20 extreme fear - 20–40 fear - 40–60 neutral - 60–80 greed - 80–100 extreme greed 50 means zero net flow: above it net inflows dominate (risk-on), below it net outflows do (risk-off). 3. Use `components` to say what drives the reading — for example a high trend with a lower flow means the last month was strong but today is quieter; low breadth means a few large funds carry the total. 4. Add the evidence behind the reading: call `get_weekly_analytics` and say whether the last week's flows agree with the index. If they seem to disagree, say so plainly: the trend component looks back 20 trading days, so the index moves more slowly than a single week of flows. 5. Answer in two or three sentences: the value and zone, what it means, and the supporting flow picture. End with "Data: cryptoetf.today" and link https://cryptoetf.today/en/crypto-etf-sentiment-index for the chart and history. ## Rules - The connector returns only the current value and its components. Don't invent past values or per-asset CEFI scores. For history, point to the index page. - Don't describe weights or a formula for the index beyond what is written here. - Treat CEFI as a description of ETF demand, not a trading signal. Don't give investment advice or price predictions.
etf-flow-history3.53 KB
---
name: etf-flow-history
description: Analyze the recent flow history of a crypto ETF asset or fund, or compare several. Use when the user asks about the trend, streaks or momentum of Bitcoin, Ethereum, Solana, XRP or other crypto ETF flows over recent weeks, wants two assets or two funds (such as IBIT and FBTC) compared, or asks for a chart of daily ETF flows.
---
# Crypto ETF flow history and comparison
Read up to 30 days of daily net flows for one or more assets and explain the trend.
## Steps
1. Map what the user named to an asset code: `btc`, `eth`, `sol`, `xrp`, `hyp` (Hyperliquid / HYPE), `doge`, `link` (Chainlink), `avax` (Avalanche), `hbar` (Hedera), `ltc` (Litecoin), `bnb`, `dot` (Polkadot), `sui`, `near`, `trx` (TRON), `zec` (Zcash). If they named something else, say it isn't tracked and list the sixteen assets.
2. Call `get_asset_flows` once per asset. Each call returns `days`: a list of `{date, netFlowUsdM}` in ascending date order. If the question is about funds (IBIT vs FBTC, which issuer is gaining share), call `get_fund_flows` for the asset instead and read each fund's series from its `days`.
3. Drop non-trading days before computing anything (see Data rules), then compute what the question needs:
- net total for the period, and for comparisons the same period for every asset;
- the current streak of inflow or outflow days and the longest streak in the window;
- the largest inflow day and the largest outflow day;
- the last 5 trading days against the 5 before them, to call momentum improving or fading.
4. Write the answer: one-sentence verdict first, then the supporting numbers. For a comparison, a small table with one row per asset.
5. If the user wants a chart and you can make one, plot daily net flow as bars (green for inflow, red for outflow) over trading days.
6. End with "Data: cryptoetf.today", and link the asset page for more detail, for example https://cryptoetf.today/en/ethereum-etf-flows.
## Data rules
- Figures are **net USD millions**; positive is an inflow.
- **Weekends and US market holidays are not trading days.** Some assets list them with `netFlowUsdM: 0` and others omit them. Remove Saturday and Sunday dates before you count streaks, averages or "flat days", or the numbers will be wrong.
- A weekday zero in a small asset usually means no creations or redemptions, not a neutral signal.
- History covers the **last 30 days** only. For longer history, point to the asset page on cryptoetf.today rather than guessing.
- The most recent day is normally yesterday; flows are reported one business day in arrears.
- `get_fund_flows` breaks one asset down by fund (IBIT, FBTC, ETHA and the rest) for the last 30 days: each day lists `ticker`, `issuer` and `netFlowUsdM` for every fund trading that day, plus `totalUsdM`. `netFlowUsdM: null` means there is no figure for the fund that day yet: say it hasn't reported, don't count it as zero. Current net assets (AUM), coins held, sponsor fees and listing dates come from `get_funds`: one snapshot with its date, not a history. `coinsHeld` is null for issuers that don't publish holdings in coins (Fidelity, Grayscale, Franklin Templeton, Canary, WisdomTree); don't fill it in from the price. Never invent a fund figure that isn't in a response. `get_fund_flows` lists trading days only. For longer history, send the user to `https://cryptoetf.today/en/<slug>-etf-flows` (slugs: bitcoin, ethereum, solana, xrp, hype, dogecoin, chainlink, avalanche, hedera, litecoin, bnb, polkadot, sui, near, tron, zcash).
- Describe the flows; don't give investment advice or price predictions.
etf-flows-daily3.59 KB
---
name: etf-flows-daily
description: Brief on the latest spot crypto ETF flows. Use when the user asks what happened with Bitcoin, Ethereum or other crypto ETFs today or yesterday, whether money is flowing into or out of crypto ETFs, wants a quick market snapshot of ETF inflows and outflows, or asks which fund (IBIT, FBTC, GBTC, ETHA and others) led the flows.
---
# Daily crypto ETF flow brief
Produce a short, accurate brief on the most recent published day of US spot crypto ETF flows.
## Steps
1. Call `get_flows_summary` from the CryptoETF connector.
2. If the user asked about a specific fund or which fund led the flows, call `get_fund_flows` for that fund's asset and read the day that matches `referenceDate` (see Data rules). Don't suggest other websites or a web search for fund data.
3. If the user asked how big a fund or an asset's ETFs are, what a fund charges or who issues it, call `get_funds`. If they asked about prices or market context, also call `get_prices`. If they asked about sentiment or mood, also call `get_cefi_index`.
4. Write the brief in this order:
- **Headline**: the total net flow across all assets (`totalUsdM`) and the day it belongs to (`referenceDate`), for example "Net +$68.5M into crypto ETFs on Mon, Sep 28".
- **Leaders**: the assets with the largest inflow and the largest outflow, with their figures. If you fetched `get_fund_flows`, name the funds behind the move.
- **The rest**: one line for the remaining assets that moved; group all assets at zero into a single phrase ("no flows in HYPE, DOGE, AVAX…").
- **Context** (only if you fetched it): price moves from `get_prices`, CEFI level from `get_cefi_index`.
5. End with the source line: "Data: cryptoetf.today".
Keep the brief under 150 words unless the user asks for more. Use a table only when the user asks for all assets.
## Data rules
- Figures are **net USD millions**. Positive is an inflow, negative an outflow. Write them as `+$31.0M` / `−$12.4M`; convert to billions above 1,000.
- Flows are published **one business day in arrears**. `referenceIsToday` is normally `false`: never call the figures "today's". Name the actual date. On a Monday the freshest data is Friday's.
- Each asset has its own `date`. If an asset's date differs from `referenceDate`, say its figure is from that earlier day instead of mixing days silently.
- `HYP` is Hyperliquid (token ticker HYPE).
- A zero for a smaller asset (DOGE, AVAX, HBAR, BNB, DOT, SUI, TRX and others) usually means no creations or redemptions that day, which is normal for young, small funds. Don't describe it as a signal.
- `get_fund_flows` breaks one asset down by fund (IBIT, FBTC, ETHA and the rest) for the last 30 days: each day lists `ticker`, `issuer` and `netFlowUsdM` for every fund trading that day, plus `totalUsdM`. `netFlowUsdM: null` means there is no figure for the fund that day yet: say it hasn't reported, don't count it as zero. Current net assets (AUM), coins held, sponsor fees and listing dates come from `get_funds`: one snapshot with its date, not a history. `coinsHeld` is null for issuers that don't publish holdings in coins (Fidelity, Grayscale, Franklin Templeton, Canary, WisdomTree); don't fill it in from the price. Never invent a fund figure that isn't in a response. For history beyond 30 days, send the user to the asset's page: `https://cryptoetf.today/en/<slug>-etf-flows`, where the slug is bitcoin, ethereum, solana, xrp, hype, dogecoin, chainlink, avalanche, hedera, litecoin, bnb, polkadot, sui, near, tron or zcash.
- Describe what the flows show. Don't give buy, sell or price-target advice; this is market data, not investment advice.
etf-flows-weekly2.75 KB
---
name: etf-flows-weekly
description: Weekly recap of spot crypto ETF flows. Use when the user asks how crypto ETFs did this week or over the last seven days, which asset had the strongest or weakest week of inflows, or wants a weekly market wrap-up for a newsletter, post or report.
---
# Weekly crypto ETF flow recap
Summarize the last seven days of US spot crypto ETF flows across all tracked assets.
## Steps
1. Call `get_weekly_analytics` from the CryptoETF connector. It returns, per asset, the net flow for the window (`netFlowUsdM`), `avgDailyUsdM`, and the number of positive and negative days, plus the window (`from`, `to`).
2. If the user wants the day-by-day shape for a major asset (for example "was it one big day or steady buying?"), call `get_asset_flows` for that asset and read the days inside the window.
3. Write the recap:
- **Headline**: the combined net flow of all assets for the window, with the dates.
- **Ranking**: assets ordered by net flow, largest inflow first. Show the top movers with their positive/negative day counts ("BTC +$703.1M, 4 of 4 trading days positive").
- **Breadth**: how many assets saw net inflows versus net outflows.
- **Notable**: one or two observations, such as a streak, a reversal from outflows, or a single day that made most of the week.
4. End with "Data: cryptoetf.today".
Offer a table when more than five assets moved or when the user wants something to paste into a report.
## Data rules
- Figures are **net USD millions**; positive is an inflow.
- The window covers the last seven calendar days. Use `tradingDays` for the number of trading days in it, and `avgDailyUsdM` as the average per trading day. If a response has no `tradingDays` field, don't quote `avgDailyUsdM`: in that older format it isn't comparable across assets. Report the net flow and day counts instead.
- Positive plus negative days is the number of days with non-zero flow. Trading days beyond that had no creations or redemptions, which is common for small funds.
- The latest published day is normally yesterday: flows are reported one business day in arrears.
- `HYP` is Hyperliquid (token ticker HYPE).
- `get_weekly_analytics` gives asset-level totals. For a fund's week, call `get_fund_flows` for its asset and add up that fund's `netFlowUsdM` over the days inside `from`–`to`; a `null` day means the fund hasn't reported, so say the total is partial. Current net assets (AUM), coins held, sponsor fees and listing dates come from `get_funds`: one snapshot with its date, not a history. `coinsHeld` is null for issuers that don't publish holdings in coins (Fidelity, Grayscale, Franklin Templeton, Canary, WisdomTree); don't fill it in from the price. Don't invent fund figures.
- Describe the flows; don't give investment advice or price predictions.
etf-fund-metrics5.84 KB
---
name: etf-fund-metrics
description: What sits inside US spot crypto ETFs beyond daily flows. Use when the user asks how many coins the ETFs hold or what share of supply they own, whether ETF holders are in profit or underwater and at what average entry price, whether bitcoin ETFs buy more than miners produce, which funds stake and at what yield, or how much issuers earn from fees.
---
# Inside the funds: holdings, entry price, staking, fees
Five connector tools describe the stock of the funds rather than one day's flow. Pick the one that matches the question; don't call all five.
| Question | Tool | Page for charts and history |
|---|---|---|
| How many coins do the ETFs hold? What share of supply? | `get_etf_holdings` | https://cryptoetf.today/en/crypto-etf-holdings (bitcoin: /en/bitcoin-etf-holdings) |
| Are ETF holders in profit? What is their average entry? | `get_cost_basis` | https://cryptoetf.today/en/crypto-etf-cost-basis |
| Do bitcoin ETFs absorb more than miners produce? | `get_btc_etf_vs_miners` | https://cryptoetf.today/en/bitcoin-etf-buying-vs-mined |
| Which funds stake, how much, at what yield? | `get_staking` | https://cryptoetf.today/en/crypto-etf-staking (also /en/ethereum-etf-staking, /en/solana-etf-staking) |
| How much do issuers earn from fees? | `get_issuer_revenue` | https://cryptoetf.today/en/crypto-etf-issuer-revenue |
All tools except `get_btc_etf_vs_miners` take an optional `asset` (btc, eth, sol, xrp, hyp, doge, link, avax, hbar, ltc, bnb, dot, sui, near, trx, zec). Pass it when the user names one asset.
## Steps
1. Call the one tool that answers the question, filtered by asset when one is named.
2. Lead with the answer in one sentence, with the `asOf` date written out ("as of 6 November").
3. Add two or three supporting facts from the same response: the change over 7 or 30 days, the largest funds, the streak.
4. Say which figures are estimates where the response marks them (see below).
5. End with "Data: cryptoetf.today" and the page from the table above for charts and longer history.
## Reading each response
**`get_etf_holdings`**
- `coins` is the total held by all ETFs of the asset; `supplyPct` is already a percent of circulating supply (4.74 means 4.74%, not 474%).
- `coinsReported` comes from issuers; `coinsEstimated` is net assets divided by the same-day coin price for issuers that don't publish coins. Per fund, `coinsBasis` says which. When estimated coins are a large part of the total, say the total is partly estimated.
- `change7dCoins` and `change30dCoins` are changes in coins, not dollars. `change7dPctPoints` is a change in percentage points of supply.
- `history` gives the last 30 trading days of `coins` and `supplyPct` per asset.
- Converted trusts (GBTC, ETHE) are included in holdings.
**`get_cost_basis`**
- `avgEntryUsd` is the average price per coin paid through net creations and redemptions; `price` is the latest coin price; `pnlPct` is a fraction (−0.072 means −7.2%).
- `state` is `above` or `below` entry; `streakDays` and `streakSince` say for how long.
- Converted trusts (GBTC, ETHE) are left out of this calculation, so `qtyCoin` is smaller than the holdings total. Don't compare the two as if they should match.
- This is an estimate built from daily flows, not a figure any issuer reports. Call it "estimated average entry".
**`get_btc_etf_vs_miners`**
- Windows: `week` (last full Monday–Sunday week), `30d`, `ytd`, `all` (since 11 January 2024). Each has `etfNetBtc`, `minedBtc` and `ratio`.
- `ratio` is ETF net BTC divided by mined BTC. It is `null` when ETFs had net outflows: say the ETFs were net sellers in that window, don't report a ratio.
- `streakWeeks.length` is the run of full weeks with ETF demand above new supply; 0 means the latest full week was not one.
- `weeks` lists the last 12 weeks; a week with `partial: true` is still in progress and must not be read as a full week.
- Bitcoin only. For other assets there is no mined-supply comparison.
**`get_stakiPreview of the first 4,000 characters. Read full instructions →
Release notes
First release for ChatGPT. Daily and weekly flows for sixteen US spot crypto ETF complexes, per-fund flows, fund AUM and fees, coins held against circulating supply, holders' estimated average entry, bitcoin ETF demand against newly mined supply, staking status and yields, estimated issuer fee revenue, the CEFI sentiment index and spot prices.
Declared in the saved package. Remote tools may change independently.
Package details
Publisher declarations from the archived package. These are separate from our research and the live service's terms.
- Package license
- MIT
- Package author
- ZeroSix Studio
- Keywords
- See keywords
- Commerce declaration
- Does not support commerceThis does not establish whether access is free or paid.
- Review scenarios
- 5 positive · 3 negativeScenario counts, not test results.
Package observed Nov 21, 2026.
Technical details
- First seen
- Nov 21, 2026 · 00:00 UTC
- Last seen
- Nov 21, 2026 · 06:00 UTC
- Collection status
- Collected
plugin_asdk_app_6af176f448f08191a6810e1b121f4082
Download plugin data (JSON)Plugin details: freshly prepared. Prepared . Refreshed when plugin data changes or after 6 hours. Your account and Follow status are live.