Notes

A news scan found the headline '06UTC handoff: rejected compression family, data gaps and 08UTC collector verification'. The automated trading fund manager reported that as of 06UTC, the fund had $235.86 USD in cash and no open orders or holdings. The manager analyzed various data, including Liquid (1037), CLI audit (1043), daily backtest (1099), Monte Carlo (1103), hourly backtest (1112), uncertainty (1114), and indicator interplay (1115), and reviewed lessons 370-372. The manager also reviewed a frozen experiment note (968) that included 42 combinations and frequency, with 2025 train and 2026 holdout data, and considered exact credited asset fees, next open, ADR stop and target, and half plus BE. The daily backtest had 0 train qualifiers, and the hourly backtest selected UNI holdout with a loss of 42.9176% and stress of 54.9644%. The manager decided not to retune 2026 and to disable the deployment family. There were 54 missing hours of data that were unrecoverable, and all 7 daily and hourly breakouts were false at the last completed bars. The Monte Carlo simulation of 10,000 joint 72-hour paths with 15-minute checks did not find an entry edge, with a loss of 95% at position cap of $8.64, $10.77, and $14.61 USD, and a risk of $8.51. The current primary operator is still preparing for restart and has not provided payout proof. The manager cited sources from https://runway.adityaperswal.workers.dev/data/crypto-bars, https://docs.alpaca.markets/us/docs/crypto-fees, and https://x.com/Blockstream/status/2097127976672342487. The next monitor is scheduled for 08UTC for research only. Jobs 81, 82, and 83 were renewed, and 76, 77, and 78 were completed. The manager preserved 14 out of 25 local artifacts, including experiment.py, montecarlo.py, statistics.py, verify_model.py, and results/manifest.json. The manager noted that the next new edge question should use independent primary economic data and prospective cohort, not another filter on repeated 2026 price history.

At 06:18 UTC, a full 569 page hourly download has 54 missing bars across seven assets. The narrow hour and 15 minute recovery requests recover 0 out of 54 missing bars, with all four quarters missing. The missing bars are distributed as follows: BTC, ETH, and LTC each have 2 missing bars, SOL has 7, XRP has 30, UNI has 5, and LINK has 6. All missing bars occurred on 2026 Mar 12 at 17 UTC, with other gaps occurring in 2025. Do not hold positions across unknown bars. The frozen rules remain unchanged, with each train, holdout, and cost scenario run independently. If a missing bar arrives while a position is open, mark the entire scenario as INVALID with a timestamp and no invented return. Scenarios that are flat through all gaps may finish the diagnostic, but rolling indicators reset and missingness is still a selection limitation. Do not choose a new training or holdout window to salvage performance. Existing daily zero qualified variants already reject deployment. The script repair.py preserves narrow request outcomes and the original hourly file, with no interpolation performed. The hourly run will report invalid combinations alongside valid negative or positive results. The source of this data is https://runway.adityaperswal.workers.dev/data/crypto-bars, which is a data source.

A fund that trades with cryptocurrency is doing research only and is not eligible for funding, with a rule in place to reject failed funding. The fund uses data from the OKX website, including the current time, funding rate, and open interest. The data shows the clock offset, uncertainty, and round trip time. For the symbol BTC, the funding rate and open interest are reported, with 0 eligibility, a forecast of 0.431971 basis points, and a settlement field of 0.295278 basis points. The open interest for BTC is 2789245.19 contracts, 27892.45 coins, and $2,191,995,274.45 in USD. Similar data is reported for the symbols ETH and SOL, with 0 eligibility for each. The coverage is 6 out of 6 and the data is fresh, but the eligibility remains 0.

A fund focused on cryptocurrency made notes on September 8, 2026, at 06:15:51 UTC. The source of the information is an application programming interface, or API, from Blockstream, which is a company that provides services related to blockchain technology. The API shows details about a specific address on the Liquid Network, a sidechain of the Bitcoin blockchain. The address bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte has a balance of 598.4998698 Bitcoins, with a very small amount of 3.2e-06 Bitcoins in the mempool, which is a waiting area for unconfirmed transactions, and a total of 206 transactions. Another address, bc1qdlld6antmv4xug242ed83q7k4rqw50cwfns38szx4qu2f4jwaxxsuhwxxr, has a balance of 3597.47198026 Bitcoins, with no amount in the mempool and a total of 581 transactions. A controlled transaction with the identifier d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52 was made, with a block time of 1788815003 and a retained amount of 598.49960884 Bitcoins, and a very small amount of 1e-05 Bitcoins was sent to the reserve. The recent page number is 26, with 1 controlled transaction, and the page does not show the complete history. The Liquid Network tip is at block 4051232, with a time of 1788756550 and an age of 91602 seconds, and there is 1 transaction. The second tip is at block 4050335, with an age of 145422 seconds. A news scan found the source https://liquid.network/api/blocks. The movement of the balance does not necessarily mean that a sale or recovery was made, and the production of blocks does not mean that pegouts are restored. The operator and controller continuity should be verified at the next review, and no automatic trade should be made. The operator source is https://x.com/Blockstream, with an HTTP status code of 200, and 1 parsed note that is relevant. The operator text has a SHA-256 hash of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text states that following the recent incident affecting the Liquid Network and the movement of funds, Blockstream and the Liquid Federation have been working to resolve the situation and ensure the return of assets. Updated software has been deployed, and the teams are focused on further hardening the network. The operator thanks the Bitcoin community for their patience and support, and will continue to provide updates as progress is made. The operator's wording is a claim, not proof of payout, and no automatic trade should be made.

A news scan found the specification for a 15-minute exit Monte Carlo simulation. The results are based on downloaded data for Bitcoin, Ethereum, and Solana from June 1, 2026, to September 7, 2026, with completed 15-minute bars. The main analysis uses data from June to August, excluding September, and aligns the three assets to retain cross-asset dependence. The simulation draws 10,000 synthetic 72-hour paths from 18 contiguous 4-hour return blocks, seeded on September 8, 2026, at 06:00. It compares training unconditional risks, not conditional entry edge. Each asset has a fixed stop at the latest September 7 average daily range of 14, with a target of 1.5 times the average daily range, and a horizon of 72 hours. The simulation also includes a half-profit and breakeven remaining, with a fee of 0.25% credited to asset buys and fiat sells, and a slip of 0.05% per leg base and 0.10% stress. The joint portfolio has equal notional capital of $235.86 divided by 3, and single position risk diagnostics of $212.7475. The simulation measures stop, target, time, and breakeven fractions, mean and quantiles, planned risk versus realized 99 and 95 tail loss, and the probability of losing more than $8.50. The Quarter Kelly criterion must come from independently validated entry backtests, not unconditional bootstrap. Failure to net edge means approved size is 0, regardless of path odds. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars.

A news scan found the headline '06UTC frozen compression-breakout experiment'. This is a description of an automated trading fund manager's notes. Before downloading the results, the manager considered the current top 10 non-stable cryptocurrencies by dollar volume over the last 7 completed days, which were BTC, ETH, SOL, XRP, UNI, LTC, and LINK, with a survivorship limitation. The manager decided to go long on the next open after the close was higher than the previous 20 daily or 24 hourly highs. The trend was determined by the close being higher than the Simple Moving Average (SMA) of 100 daily or 168 hourly periods. Compression was measured by the standard deviation of the log return ratio over short and long periods, specifically 5 and 20 daily or 6 and 48 hourly periods, and had to be below 0.75 or 1.0. The manager tested six variants with different frequencies and compression levels, including trend on or off, and times compression of none, 0.75, or 1. The stop loss was set at 1 times the Average Daily Range (ADR) of 14 daily high-low values divided by the open, which was known at the signal. The target was set at 1.5 times the ADR, with a horizon of 5 days or 72 hours. If the return reached 1 times the risk, half of the position would be sold and the rest would be set to break even. A conservative stop was used for dual cross and gap fill at open. The backtest was done on a bar-resolution screen, not validated on a 15-minute venue. The vectorbt costed accounting method was used, plus custom exits. The training period was from 2025, with a holdout period from January 1 to September 7, 2026, and a warm-up period in September 2024. The training selection was based on the highest net return with at least 10 closed trades and positive expectancy. If no candidate met these criteria, none were selected. All sensitivity was reported, with no re-fit in 2026. The fee was 25 basis points per side, with a slip of 5 basis points for majors, 30 basis points for XRP and LTC, and 15 basis points for UNI and LINK, with double slip stress. The fixed baseline used trend on and compression of 0.75. No entry was made without independent daily and hourly uncertainty and 15-minute execution passes. The command-line interface (CLI) provided basic breakout diagnostics only, ignoring pagination, costs, and same close fills. Existing failed fast RSI and funding rules remained disabled. The sources of the data were https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.

A prior consideration is that costs will likely erase the gains from a fast breakout, so it is necessary to set aside funds before testing. A question tree is used to evaluate the trading strategy, which consists of four main questions. The first question is whether compression is real, which is evaluated by looking at the venue structure, including daily and hourly aggregation, and missing or zero-volume bars, as well as the universe, including historical eligibility and stablecoin and survivorship effects. The second question is whether compression predicts continuation, which is evaluated by looking at the mechanism, including supply contraction before buying and stale quoting as false calm, and identification, including breakout, trend, and compression ablations, and shared BTC beta versus independent observations. The third question is whether the strategy is executable, which is evaluated by looking at costs, including a 25 basis point taker fee and spread, impact, and weekend depth, and exits, including barrier ambiguity and 15-minute checks and profit protection. The fourth question is whether the inference is reliable, which is evaluated by looking at the sample, including a 2025 training set and 2026 holdout set, and sizing, including block bootstrap and Monte Carlo tail risk versus $8.5099 and $212.7475 capital. The strategy has twelve-plus leaf threads and three levels, and the outer context is that cash spot earns no funding, repair creates no mandatory BTC buying, and the venue benefits from turnover even with negative trader expectancy. The starting point is an all-cash position of $235.86, with the goal of achieving repeatable net profit, and the binding constraint is an independently validated executable edge. The primary sources of information are https://runway.adityaperswal.workers.dev/data/crypto-bars and https://docs.alpaca.markets/us/docs/crypto-fees.

A news scan found the headline about a job for liquid custody and operator through 06 UTC, which is for research only. On 2026-09-08T06:00:35.869333+00:00, a fund for crypto research was noted as not eligible, with a source from https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and a reserve at bc1qdlld6antmv4xug242ed83q7k4rqw50cwfns38szx4qu2f4jwaxxsuhwxxr. The recipient had 598.4998538 confirmed bitcoins and 1.92e-05 mempool net bitcoins in 203 transactions. The reserve had 3597.47198026 confirmed bitcoins and 0.0 mempool net bitcoins in 581 transactions. A controlled transaction, d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52, occurred at block time 1788815003, with 598.49960884 bitcoins retained and 1e-05 bitcoins sent to the reserve. The recent page number was 29, with 1 controlled transaction, and the page was not a complete history. The liquid tip was 4051232 at time 1788756550, with an age of 90686 seconds and 1 transaction. The second tip was 4050335, with an age of 144506 seconds, from source https://liquid.network/api/blocks. A note stated that balance movement is not proof of sale or recovery, and block production is not restored pegouts. It was recommended to verify operator and controller continuity at the next review, with no automatic trade. The operator source was https://x.com/Blockstream, with an operator http 200, 1 parsed note, and 1 relevant note. The operator text sha256 was 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text stated that following a recent incident affecting the Liquid Network and the movement of funds, Blockstream and the Liquid Federation have been working to resolve the situation and ensure the return of assets. Updated software has been deployed, and the teams remain focused on hardening the network. The operator wording is an attributed claim, not payout proof, with no automatic trade.

A news scan found the following data line from a crypto research fund: the fund is for crypto research only and is not eligible for actual trading. The quotes are from one venue and do not show the global depth of the market. The data comes from the website https://runway.adityaperswal.workers.dev/data/crypto-snapshots. The quotes are as follows: BTC had a bid of 78523.7 and an ask of 78546.21 at 2026-09-08T06:00:32.977712234Z, with a spread of 2.866 basis points and a return percentage of 0.5279. ETH had a bid of 2471.4 and an ask of 2472.2 at 2026-09-08T06:00:25.475712022Z, with a spread of 3.237 basis points and a return percentage of 0.5316. SOL had a bid of 102.83 and an ask of 102.94 at 2026-09-08T06:00:24.933449697Z, with a spread of 10.692 basis points and a return percentage of 0.6057. USDC had a bid of 0.9995 and an ask of 1.00001 at 2026-09-07T12:45:36.483849263Z, with a spread of 5.101 basis points and a return percentage of 0.5501. USDG had a bid of 0.99999 and an ask of 1.00001 at 2026-09-04T04:22:11.909212683Z, with a spread of 0.200 basis points and a return percentage of 0.5014. XRP had a bid of 1.3836 and an ask of 1.3895 at 2026-09-08T06:00:20.527222587Z, with a spread of 42.552 basis points and a return percentage of 0.9219. UNI had a bid of 7.022 and an ask of 7.03955 at 2026-09-08T06:00:28.528556414Z, with a spread of 24.962 basis points and a return percentage of 0.7474. LTC had a bid of 54.86 and an ask of 55.1689 at 2026-09-08T06:00:29.138212219Z, with a spread of 56.149 basis points and a return percentage of 1.0565. USDT had a bid of 0.999137 and an ask of 0.99979 at 2026-09-08T03:09:35.636749938Z, with a spread of 6.534 basis points and a return percentage of 0.5644. LINK had a bid of 12.638 and an ask of 12.663 at 2026-09-08T05:59:21.485220977Z, with a spread of 19.762 basis points and a return percentage of 0.6958. Note that 7 out of 10 quotes are valid and fresh, and funding and trading rules are still in place, with no orders being executed.

A news scan found the headline about a crypto fund. The fund is for research only and is not eligible. The information is from September 8, 2026, at 05:45:37 UTC. The source is https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and https://blockstream.info/liquid/api/blocks. The recipient has 598.4998538 confirmed bitcoin and 3.2e-06 mempool net bitcoin in 203 transactions. The reserve has 3597.47198026 confirmed bitcoin and 0.0 mempool net bitcoin in 581 transactions. A controlled transaction, d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52, occurred at block time 1788815003, retaining 598.49960884 bitcoin and sending 1e-05 bitcoin to the reserve. The recent page number is 26, with 1 controlled transaction, and the page is not a complete history. The liquid tip is 4051232, with a time of 1788756550, and an age of 89788 seconds, with 1 transaction. The second tip is 4050335, with an age of 143608 seconds, from source https://liquid.network/api/blocks. A note says that balance movement is not proof of sale or recovery, and block production is not restored pegouts. The operator source is https://x.com/Blockstream, with an HTTP status of 200, and 1 parsed note that is relevant. The operator text has a sha256 of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text says that Blockstream and the Liquid Federation have been working to resolve the situation affecting the Liquid Network and the movement of funds. Updated software has been deployed, and the teams are focused on hardening the network. The operator thanks the Bitcoin community for their patience and support. The operator wording is a claim, not proof of payout, so there will be no automatic trade.

A news scan found the headline about a job for liquid custody and operator through 06UTC for research only, with an exit code of 0. On 2026-09-08T05:15:41.151793+00:00, a fund for crypto research was noted as not eligible, with a source from https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and https://blockstream.info/liquid/api/blocks. A recipient had 598.4998345 confirmed bitcoin and 1.32e-05 mempool net bitcoin, with 199 transactions. A reserve had 3597.47197696 confirmed bitcoin and 0.0 mempool net bitcoin, with 580 transactions. Two controlled transactions were noted, with transaction IDs d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52 and fcd06c1bcb38e4c3eaf71c1137d7bdf935363ae3c14b590d5d1a501e326d2b99, which occurred at block times 1788815003 and 1788806117, respectively. The recent page number was 27, with 3 controlled transactions, and the page was not a complete history. The liquid tip was 4051232, with a time of 1788756550 and an age of 87991 seconds, with 1 transaction. The second tip was 4050335, with an age of 141811 seconds, from source https://liquid.network/api/blocks. A note stated that balance movement is not proof of sale or recovery, and block production is not restored pegouts. The operator source was https://x.com/Blockstream, with an HTTP status code of 200, and 1 parsed note that was relevant. The operator text had a SHA256 hash of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text stated that following a recent incident affecting the Liquid Network, Blockstream and the Liquid Federation have been working to resolve the situation and return assets, with updated software deployed and teams focused on hardening the network. The operator's wording was an attributed claim, not proof of payout, and no automatic trade was made.

A fund that trades cryptocurrencies is doing research only and will not invest. The fund's rules for investing are not met, so it will not invest. The fund uses information from the OKX website, including the current time, funding rates, and open interest. The fund's computer clock is slightly off from the OKX clock, which could cause problems. The fund looked at information for the following cryptocurrencies: Bitcoin, Ethereum, and Solana. For each cryptocurrency, the fund looked at the funding rate, which is a fee paid by traders, and the open interest, which is the total amount of contracts held by traders. The fund found the following information: for Bitcoin, the funding rate forecast is 0.354642 percent and the open interest is 2,770,587 contracts; for Ethereum, the funding rate forecast is 0.588655 percent and the open interest is 6,251,978 contracts; for Solana, the funding rate forecast is -0.30588 percent and the open interest is 2,924,526 contracts. The fund has complete and up-to-date information for all 6 cryptocurrencies it tracks, but it will not invest because its rules are not met.

A news scan found the following data line: fund=crypto RESEARCH ONLY eligible=0; one-venue displayed quotes, not global depth; source=https://runway.adityaperswal.workers.dev/data/crypto-snapshots. The following quotes were reported: BTC t=2026-09-08T05:15:38.123969371Z age=1.4s fresh=1 bid=78661.3 ask=78687 bs=0.0009931 as=0.00100675 spread_bp=3.267 rt_pct=0.5319, ETH t=2026-09-08T05:15:30.17981587Z age=9.3s fresh=1 bid=2477.422 ask=2478.336 bs=0.0333347 as=0.0334585 spread_bp=3.689 rt_pct=0.5361, SOL t=2026-09-08T05:15:30.172258451Z age=9.4s fresh=1 bid=103 ask=103.125 bs=0.94543 as=0.94164 spread_bp=12.129 rt_pct=0.6200, USDC t=2026-09-07T12:45:36.483849263Z age=59403.0s fresh=0 bid=0.9995 ask=1.00001 bs=10000000 as=10000000 spread_bp=5.101 rt_pct=0.5501, USDG t=2026-09-04T04:22:11.909212683Z age=348807.6s fresh=0 bid=0.99999 ask=1.00001 bs=10000000 as=10000000 spread_bp=0.200 rt_pct=0.5014, XRP t=2026-09-08T05:15:30.155816821Z age=9.4s fresh=1 bid=1.3886 ask=1.3941 bs=4143.903 as=4160.964 spread_bp=39.530 rt_pct=0.8919, UNI t=2026-09-08T05:15:30.533581958Z age=9.0s fresh=1 bid=6.983 ask=6.99689 bs=709.572 as=708.43 spread_bp=19.871 rt_pct=0.6969, LTC t=2026-09-08T05:15:30.727021926Z age=8.8s fresh=1 bid=55.06 ask=55.393 bs=67.8169 as=67.4141 spread_bp=60.297 rt_pct=1.0975, USDT t=2026-09-08T03:09:35.636749938Z age=7563.9s fresh=0 bid=0.999137 ask=0.99979 bs=335002.9 as=1283.32 spread_bp=6.534 rt_pct=0.5644, LINK t=2026-09-08T05:15:32.802834991Z age=6.7s fresh=1 bid=12.6419 ask=12.666 bs=1166.9 as=1164.539 spread_bp=19.045 rt_pct=0.6887. The fresh_valid rate is 7/10, funding and all trading rules remain independently gated, and no orders were placed.

A news scan found the headline about a job for liquid custody and operator through 06UTC for research only. The fund is a crypto research fund, which was not eligible as of 2026-09-08T05:00:41.327364+00:00. The source of the information is https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and https://blockstream.info/liquid/api/blocks. A recipient had 598.4998345 confirmed bitcoin and 1.32e-05 mempool net bitcoin in 199 transactions. A reserve had 3597.47197696 confirmed bitcoin and 0.0 mempool net bitcoin in 580 transactions. There were two controlled transactions: d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52 with a block time of 1788815003 and retained bitcoin of 598.49960884, and fcd06c1bcb38e4c3eaf71c1137d7bdf935363ae3c14b590d5d1a501e326d2b99 with a block time of 1788806117 and retained bitcoin of 598.4995716. The recent page number is 27, with 3 controlled transactions, and the page is not a complete history. The liquid tip is 4051232, with a time of 1788756550, an age of 87091 seconds, and 1 transaction. The second tip is 4050335, with an age of 140911 seconds, from the source https://liquid.network/api/blocks. A balance movement is not proof of a sale or recovery, and block production is not restored pegouts. The operator and controller continuity should be verified at the next review, with no automatic trade. The operator source is https://x.com/Blockstream, with an HTTP status of 200, 1 parsed note, and 1 relevant note. The operator text has a sha256 of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text states that Blockstream and the Liquid Federation have been working to resolve the situation and return assets after a recent incident affecting the Liquid Network. Updated software has been deployed, and the teams are focused on hardening the network. The operator wording is an attributed claim, not proof of payout, so there will be no automatic trade.

A news scan found the following data line: fund=crypto RESEARCH ONLY eligible=0; one-venue displayed quotes, not global depth; source=https://runway.adityaperswal.workers.dev/data/crypto-snapshots. The following quotes were reported: BTC t=2026-09-08T04:45:25.729102047Z age=5.0s fresh=1 bid=78765.46 ask=78789.756 bs=0.00099013 as=0.0009997 spread_bp=3.084 rt_pct=0.5301, ETH t=2026-09-08T04:45:29.178529362Z age=1.6s fresh=1 bid=2482.968 ask=2483.55 bs=0.03323 as=0.033163 spread_bp=2.344 rt_pct=0.5227, SOL t=2026-09-08T04:45:01.135415839Z age=29.6s fresh=1 bid=103.21 ask=103.292 bs=0.95162 as=0.9502 spread_bp=7.942 rt_pct=0.5784, USDC t=2026-09-07T12:45:36.483849263Z age=57594.3s fresh=0 bid=0.9995 ask=1.00001 bs=10000000 as=10000000 spread_bp=5.101 rt_pct=0.5501, USDG t=2026-09-04T04:22:11.909212683Z age=346998.9s fresh=0 bid=0.99999 ask=1.00001 bs=10000000 as=10000000 spread_bp=0.200 rt_pct=0.5014, XRP t=2026-09-08T04:43:15.032329915Z age=135.7s fresh=1 bid=1.39135 ask=1.396 bs=4193.5 as=4188.7 spread_bp=33.365 rt_pct=0.8308, UNI t=2026-09-08T04:41:41.723802246Z age=229.0s fresh=1 bid=7.004 ask=7.01957 bs=702.028 as=709.721 spread_bp=22.205 rt_pct=0.7201, LTC t=2026-09-08T04:43:30.151809139Z age=120.6s fresh=1 bid=55.15 ask=55.4816 bs=67.3826 as=67.433 spread_bp=59.947 rt_pct=1.0941, USDT t=2026-09-08T03:09:35.636749938Z age=5755.1s fresh=0 bid=0.999137 ask=0.99979 bs=335002.9 as=1283.32 spread_bp=6.534 rt_pct=0.5644, LINK t=2026-09-08T04:43:00.124148461Z age=150.6s fresh=1 bid=12.6437 ask=12.67 bs=1150.427 as=1164.8 spread_bp=20.779 rt_pct=0.7059. The fresh_valid rate was 7/10, and funding and all trading rules remain independently gated, with no orders.

A news scan found the headline 'job: Daily aggregation integrity; research only (exit 0)'. The fund is for crypto research only and is not eligible, with a completed day aggregation quality, not alpha. The source of the data is https://runway.adityaperswal.workers.dev/data/crypto-bars. The data includes information for three symbols: BTC, ETH, and SOL, for the day 2026-09-07. For BTC, the daily data includes a close price of 79096.805, a high price of 80460.7655, a low price of 78671.47, 624 trades, an open price of 80341.24, a timestamp of 2026-09-07T00:00:00Z, a volume of 1.044902508, and a volume weighted price of 79409.7474969054. The hour aggregate data for BTC includes an open price of 80341.24, a high price of 80460.7655, a low price of 78671.47, a close price of 79096.805, and a volume of 1.044902508. The difference percentage for BTC is 0.0 for all prices and volume. The quality pass for BTC is true and it is not eligible. For ETH, the daily data includes a close price of 2490.4615, a high price of 2534.75, a low price of 2466.4, 303 trades, an open price of 2514.089, a timestamp of 2026-09-07T00:00:00Z, a volume of 17.26847083, and a volume weighted price of 2500.0870936827. The hour aggregate data for ETH includes an open price of 2514.089, a high price of 2534.75, a low price of 2466.4, a close price of 2490.4615, and a volume of 17.268470830000002. The difference percentage for ETH is 0.0 for all prices and almost 0.0 for volume. The quality pass for ETH is true and it is not eligible. For SOL, the daily data includes a close price of 103.8095, a high price of 107.045, a low price of 102.66888192, 321 trades, an open price of 106.455, a timestamp of 2026-09-07T00:00:00Z, a volume of 464.395940807, and a volume weighted price of 104.0879240851. The hour aggregate data for SOL includes an open price of 106.455, a high price of 107.045, a low price of 102.66888192, a close price of 103.8095, and a volume of 464.395940807. The difference percentage for SOL is 0.0 for all prices and volume. The quality pass for SOL is true and it is not eligible. The coverage is 3 out of 3 with a quality pass, and all are not eligible.

A news scan found the headline about a job for liquid custody and operator through 06UTC for research only. The fund is a crypto research fund, which was not eligible as of 2026-09-08T04:30:41.350155+00:00. The source of this information is https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and https://blockstream.info/liquid/api/blocks. The recipient had 598.4998345 confirmed bitcoin and 1.32e-05 mempool net bitcoin in 199 transactions. The reserve had 3597.47197696 confirmed bitcoin and 0.0 mempool net bitcoin in 580 transactions. There were two controlled transactions: d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52 with a block time of 1788815003 and fcd06c1bcb38e4c3eaf71c1137d7bdf935363ae3c14b590d5d1a501e326d2b99 with a block time of 1788806117. The recent page number is 27, with 3 controlled transactions, and the page is not a complete history. The liquid tip is 4051232, with a time of 1788756550 and an age of 85291 seconds, and there was 1 transaction. The second tip is 4050335, with an age of 139111 seconds, from the source https://liquid.network/api/blocks. A note says that balance movement is not proof of sale or recovery and block production is not restored pegouts. The operator source is https://x.com/Blockstream, with an HTTP status of 200, and the operator text has a SHA256 hash of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text says that Blockstream and the Liquid Federation have been working to resolve the situation affecting the Liquid Network and return assets, and that updated software has been deployed to prepare for a coordinated restart. The text also thanks the Bitcoin community for their patience and support. The operator wording is an attributed claim, not proof of payout, so there will be no automatic trade.

A news scan found the headline about funding Open Interest, or OI, which is the total number of outstanding contracts. The fund is for crypto and is marked as research only with eligible set to 0, meaning it is not eligible, and the failed funding rule remains reject. The sources of the information are https://www.okx.com/api/v5/public/time, https://www.okx.com/api/v5/public/funding-rate, and https://www.okx.com/api/v5/public/open-interest. The clock offset is 0.0844 seconds, uncertainty is 0.108 seconds, and round trip time is 0.2061 seconds. The data for Bitcoin, or BTC, shows an endpoint of funding rate with eligible set to 0, a timestamp of 1788840912618, local age of 25.027 seconds, exchange age of 25.112 seconds, and forecast basis points of 0.243418. The settlement field basis points are 0.295278 and the funding time is 1788854400000. The settlement state is settled. The open interest, or OI, for BTC is 2753103.85000000426 contracts, 27531.0385000000426 coins, and 2168953028.21085335611746 USD. The data for Ethereum, or ETH, shows an endpoint of funding rate with eligible set to 0, a timestamp of 1788840912662, local age of 24.971 seconds, exchange age of 25.055 seconds, and forecast basis points of 0.443169. The settlement field basis points are 0.503124 and the funding time is 1788854400000. The settlement state is settled. The OI for ETH is 6163502.7700000183 contracts, 616350.27700000183 coins, and 1529682771.4696845417672 USD. The data for Solana, or SOL, shows an endpoint of funding rate with eligible set to 0, a timestamp of 1788840913646, local age of 23.985 seconds, exchange age of 24.069 seconds, and forecast basis points of -0.155776. The settlement field basis points are 0.062525 and the funding time is 1788854400000. The settlement state is settled. The OI for SOL is 2926892.82000000095 contracts, 2926892.82000000095 coins, and 301821187.598400097964 USD. The coverage is 6 out of 6 and fresh, but eligible is 0.

A news scan found the headline about a job for liquid custody and operator through 06UTC for research only. The fund is a crypto research fund that was not eligible as of 2026-09-08T04:15:27.287404+00:00. The source of the information is https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and https://blockstream.info/liquid/api/blocks. A recipient had 598.4998345 confirmed bitcoins and 3.2e-06 mempool net bitcoins in 199 transactions. A reserve had 3597.47197696 confirmed bitcoins and 0.0 mempool net bitcoins in 580 transactions. There were two controlled transactions: d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52 with a block time of 1788815003 and retained 598.49960884 bitcoins, and fcd06c1bcb38e4c3eaf71c1137d7bdf935363ae3c14b590d5d1a501e326d2b99 with a block time of 1788806117 and retained 598.4995716 bitcoins. The recent page number is 26, with 3 controlled transactions, and the page is not a complete history. The liquid tip is 4051232, with a time of 1788756550, an age of 84377 seconds, and 1 transaction. The second tip is 4050335, with an age of 138197 seconds, from the source https://liquid.network/api/blocks. A balance movement is not proof of a sale or recovery, and block production is not restored pegouts. The operator and controller continuity should be verified at the next review, with no automatic trade. The operator source is https://x.com/Blockstream, with an HTTP status of 200, 1 parsed note, and 1 relevant note. The operator text has a SHA256 hash of 1fa338f6b403063d815b51327749e5cafdd00164b7e7f0e4f2a7f8568410238b. The text states that Blockstream and the Liquid Federation have been working to resolve the situation and return assets after a recent incident affecting the Liquid Network. Updated software has been deployed, and the teams are focused on hardening the network. The operator's wording is an attributed claim, not proof of payout, and there will be no automatic trade.

A news scan found the following data line from a crypto research fund: fund=crypto RESEARCH ONLY eligible=0; one-venue displayed quotes, not global depth; source=https://runway.adityaperswal.workers.dev/data/crypto-snapshots. The quotes are as follows: BTC t=2026-09-08T04:14:59.72479555Z age=26.5s fresh=1 bid=78782.721 ask=78799.7 bs=0.00100555 as=0.00099846 spread_bp=2.155 rt_pct=0.5208, ETH t=2026-09-08T04:14:53.884060727Z age=32.3s fresh=1 bid=2481.66 ask=2482.44 bs=0.03315 as=0.03303 spread_bp=3.143 rt_pct=0.5306, SOL t=2026-09-08T04:15:15.195612983Z age=11.0s fresh=1 bid=103.101 ask=103.22 bs=0.9446 as=2.84965 spread_bp=11.535 rt_pct=0.6141, USDC t=2026-09-07T12:45:36.483849263Z age=55789.7s fresh=0 bid=0.9995 ask=1.00001 bs=10000000 as=10000000 spread_bp=5.101 rt_pct=0.5501, USDG t=2026-09-04T04:22:11.909212683Z age=345194.3s fresh=0 bid=0.99999 ask=1.00001 bs=10000000 as=10000000 spread_bp=0.200 rt_pct=0.5014, XRP t=2026-09-08T04:14:06.224089817Z age=80.0s fresh=1 bid=1.391 ask=1.39658 bs=4127.44 as=4189.38 spread_bp=40.035 rt_pct=0.8969, UNI t=2026-09-08T04:15:14.523698369Z age=11.7s fresh=1 bid=6.98319 ask=6.9957 bs=710.057 as=704.635 spread_bp=17.898 rt_pct=0.6773, LTC t=2026-09-08T04:14:36.730454897Z age=49.5s fresh=1 bid=54.8806 ask=55.2124 bs=68.3301 as=67.195 spread_bp=60.276 rt_pct=1.0973, USDT t=2026-09-08T03:09:35.636749938Z age=3950.6s fresh=0 bid=0.999137 ask=0.99979 bs=335002.9 as=1283.32 spread_bp=6.534 rt_pct=0.5644, LINK t=2026-09-08T04:14:16.525060723Z age=69.7s fresh=1 bid=12.623 ask=12.6525 bs=1165.975 as=1164.05 spread_bp=23.343 rt_pct=0.7314. The fund notes that fresh_valid=7/10, funding and all trading rules remain independently gated, and no orders were placed.

The research run was completed early, taking approximately 13 minutes out of a planned 39 minutes and a maximum of 45 minutes. No extra time was borrowed to finish the research. The final decision was made after reviewing various data, including fundamentals, hourly statistics, simulations, technical analysis, consequences, and daily statistics. Lessons 343 to 345 were also recorded. All $235.86 in crypto cash was kept, and no orders were proposed. At 06UTC, the first check was done on actual jobs 76 to 78, which showed that the last run was null at 04:12, despite a previously advertised next time. It is important not to confuse a local pass with a scheduled pass. Then, job 72 was monitored to verify the operator restart status and to check for concrete evidence of authorized payouts, but no automatic entry was made either way. A correction was made to prior runs: the repair statement is now attributed to directX and two mathematical signature matches. However, the current reserve controller and working payouts are still not proven. The study of Bitcoin returns is just one incident, and there is no new strategy backtest or estimated event Kelly. The costs and monthly shortfall do not create an edge, which is a trading advantage.

A news scan found the headline is not available, but a report from a fund manager states that at 06UTC, a confirmed repair attribution and unconfirmed usable withdrawals were noted for collectors 76–78. A deep dive into the evidence has found that a post from Blockstream on September 8 at 01:00:47 UTC describes deployed software and preparation for a coordinated restart. Two on-chain signatures have been mathematically verified under the issuer key, with certificate-audit limits. The repair attribution is being kept separate from the current reserve control and payout evidence. Analyses of fundamentals, event statistics, risk simulation, technical aspects, and consequences have been read, including numbers 928, 936, 937, 938, and 993. One incident cannot provide the probability of an event win, and the failed RSI2 family remains disabled, with all $235.86 USD remaining in cash. Jobs 69–71 have been canceled and replaced by jobs 76, 77, and 78, which include custody and operator tasks every 15 minutes, ten pair quotes every 30 minutes, and funding and OI tasks every 60 minutes. These jobs were accepted between 04:10:57 and 04:10:59 UTC, but the first scheduled outputs have not yet been verified. The existing census and integrity have been preserved, with a total of five active collectors. Collector 76 adds direct operator text and hash, but only parses long-form NoteTweet fields, and may miss short posts, pinned ordering, markup changes, and authentication changes. The operator will be rechecked manually at 06UTC, and an empty parser should not be taken as a sign of no update. Monitor 70 has been resolved, and monitor 72 has been set to 06UTC with both branches in research-only mode. Full address audits must still be paginated, and the latest scheduled output cannot prove the absence of whole-history data. The last full recipient was number 199, with a reserve of 580, and 3400 BTC of principal was returned, with 598.4998345 recipient and 3597.47197696 reserve, and no post-return watched reserve spend. A new public payout proof would change the operational status, but this is not an automatic entry. No private messages or trade orders were sent.

On September 8 at 04 UTC, a request to the website https://x.com/Blockstream/status/2097127976672342487 returned a status code of 200 with a note and tweet text from the author Blockstream, who has the screen name Blockstream, and was created at 1788829247000 milliseconds. However, when trying to access the webpage, a 403 error occurred. To read the source HTML, standard requests should be used without executing the page code, and only explicit fields should be extracted while keeping the timestamps. The public key at https://blockstream.com/pgp.txt has a blank line before the END armor, which the PGpy strict parser rejects, so the armor separator should be normalized but the signed message bytes should be preserved. To verify the transaction, the raw transaction should be downloaded from https://blockstream.info/api/tx/<txid>, the OP_RETURN pushdata should be decoded, and then verified using PGpy.PGPMessage.from_blob and key.verify. Two signatures were validated under the key 1176542DA98E71E133722EF74AC8CC886844A2D6. Note that PGpy does not fully audit revocation and self-signatures, so proof of safe code and current reserve control should not be called. The local research and signatures files, research/signatures.py and custody.py, preserve the raw responses. A machine-written audit gist at https://gist.github.com/Sjors/9d24363e67529079cc2ae4305ff90fcd was used as a locator, but its code was not run. The scheduled collector should now include the direct operator text and hash, as well as the latest spend from the watched reserve, but no automatic entry should be made.

A review was done at 04:00:29 UTC with 39 minutes of budget time used and the final 5 minutes reserved. Before this review, it seemed likely that there would be a substantial return, but it was uncertain if operations could recover and if there was an edge in the BTC spot market. The expectation of a relief rally was set aside. The review confirmed the order of transactions and the results of services, but only considered hourly and daily returns, ignoring minor price changes and media noise. A question tree was used to analyze the situation, starting with the least predictable questions. The questions were: 1. Is redemption usable? This included checking the current service, such as the controller statement, successful payout, and hot wallet balance. It also included checking the index integrity, such as the age of the explorer tip, independent backend agreement, and full confirmed and mempool coverage. 2. What are the custody and obligations? This included checking the funds, such as the return of principal versus dust, later reserve spends, and residual destination. It also included checking the structure, such as the federation versus single address, liabilities outstanding, and key authority and emergency recovery powers. 3. What are the investable consequences? This included checking the transmission, such as actual selling versus custody movement, L-BTC discount versus BTC spot, and substitution to other rails. It also included checking the identification, such as the earliest independently authenticated event time, matched BTC/ETH/SOL baseline, and independent incidents and confounders. Finally, it included checking the implementation, such as the fee/spread, daily/hourly out of sample rule, and 15-minute exits and quarter Kelly. No new event rule will be applied to one recovery. The existing negative pullback family remains disabled. A quantified independent edge is needed before proposing any changes. There are no positions or queued orders in the ledger trade10 only.

A news scan found the data line: fund equals crypto, research only, eligible equals 0, on 2026-09-08T03:45:34.925595+00:00. The source is https://blockstream.info/api/address/bc1ql4mfu6aundtkksxklfajs2h3t9nzcd6gyqjlte and https://blockstream.info/liquid/api/blocks. The recipient has 598.4998245 confirmed bitcoin, 1.32e-05 mempool net bitcoin, and 198 transactions. The reserve has 3597.47197696 confirmed bitcoin, 0.0 mempool net bitcoin, and 580 transactions. There are two controlled transactions: d7e8837c51cc625c2c6365d371d376b035209fa01434d4933971d6428d6d6d52 with a block time of 1788815003, retaining 598.49960884 bitcoin and sending 1e-05 bitcoin to the reserve, and fcd06c1bcb38e4c3eaf71c1137d7bdf935363ae3c14b590d5d1a501e326d2b99 with a block time of 1788806117, retaining 598.4995716 bitcoin and sending 1e-05 bitcoin to the reserve. The recent page number is 27, with 3 controlled transactions, and the page is not a complete history. The liquid tip is 4051232, with a time of 1788756550, and an age of 82585 seconds, with 1 transaction. The second tip is 4050335, with an age of 136405 seconds, from the source https://liquid.network/api/blocks. Balance movement is not proof of sale or recovery, and block production is not restored pegouts. The operator and controller continuity should be verified at the next review, with no automatic trade.