Cardano’s epoch 660, which began at 06:44 JST on October 7, carried 1,083,871 transactions in its first three days. That is 6.86 times the 158,011 transactions of the whole previous epoch, 659, which ran for five days, and the average number of transactions per block rose from 7.45 to 84.0.
Most of them were transactions sending exactly 2 ADA to one and the same address: a campaign HOSKY announced right as the epoch began, in which sending 2 ADA brings HOSKY tokens back. The blocks we sampled on the morning of October 10 were each filled to about 97% of the size limit. Fees reached 4.15 times the whole of epoch 659, and that amount is added to the pot that funds staking rewards.
In the same week, governance moved toward an early-week deadline. The action to lower minPoolCost to 75 ada and the action to withdraw 11,787,063 ada for the OpenZeppelin tool stack stop accepting votes at 06:44:51 JST on October 12. In the SPO poll on raising k to 1000, explicit no votes grew to 26 pools and 1,005,266,980 ada.
This week SIPO’s three pools voted No on that k poll, and SIPO’s DRep abstained. SIPO’s first pool sits above the k=1000 saturation line, and that position is disclosed in Chapter 4.
Executive Summary
- Epoch 660 logged 1.08 million transactions in three days: as of 07:16 JST on October 10, 12,897 blocks and 1,083,871 transactions. That is 6.86 times the whole previous epoch; at this pace the five-day epoch would reach about 1.79 million. Fees were 205,533 ada, 4.15 times the whole previous epoch.
- The count was driven by HOSKY’s “Doggie Bowl”: at 06:45 JST on October 7, HOSKY announced a campaign: “Send 2 ADA → get 1.XX ADA back + a random amount of $HOSKY.” In three blocks sampled on the morning of October 10, 136 of the first 180 transactions sent exactly 2 ADA to the same address. The blocks were about 97% of the 90,112-byte limit.
- The k poll now has 26 pools explicitly against: nine pools voted No this week, five of them with a rationale authored by the Cardano Foundation, which says “More pool IDs are not more operators.” By delegated stake, No now outweighs Yes 14.79 times (9.08 times last issue).
- SIPO voted No with three pools and abstained as a DRep: recorded between 09:47 and 09:52 JST on October 7. The reason: “a change that re-divides rewards without changing their total does not grow Cardano.”
- minPoolCost’s approval rate went from 17.42% to 34.45%: Yes votes went from 48 to 76, adding about 620 million ada of voting power. To reach 67%, 67.85% of the stake that has not voted would still have to vote Yes.
- OpenZeppelin would reach only 50.28% even if every non-voter voted Yes: cast No votes are 11.94 times the Yes. The deadline is the same 06:44:51 JST on October 12. The Constitutional Committee stays at three Yes on both, with no new votes this week.
- Midnight node 1.0.400: a release that includes ledger 8.1.3, fixing an advisory rated “Critical,” and states the fix was already running on three networks, including mainnet. The reviewed deployment-request process was also formally closed.
- CIP-113 went live on mainnet: announced by the Cardano Foundation at TOKEN2049 on October 7. It lets rules such as freezing and transfer restrictions be enforced by the ledger while the token stays a native asset.
- CAP-12 has a revised draft: a constitutional amendment adding the new Dijkstra parameters to the Constitution’s appendix; the October 9 revision gave the minimum pool margin a definition and guardrails. SIPO commented on the draft on October 3.
- 30% of blocks still come from older nodes: on October 9, Intersect reported that 30% of epoch 660 blocks were produced on versions older than 11.1.2, and it continues to urge upgrades.
- Blockfrost is closing IceBreakers: the program in which SPOs served API responses, with more than 100 operators taking part.
- Markets: ETH fell 6.62%: ADA +0.12%, BTC −2.25%. NIGHT was measured from different sources at the two ends of the week, so we do not give a single weekly change.
1. Market Pulse
Values from our fixed morning observation, October 3 to October 10.
| Asset | 10/3 | 10/10 | Week |
|---|---|---|---|
| ADA | $0.239712 | $0.24 | +0.12% |
| BTC | $84,427 | $82,527 | −2.25% |
| ETH | $2,658.23 | $2,482.30 | −6.62% |
| NIGHT | $0.04617263 | $0.045657 | (not shown: different sources) |
ETH fell the most over the week, and BTC fell a little over 2%. ADA ended the week at almost the same level it started.
There was a peak in between, though. At the end of epoch 659, 06:44 JST on October 7, ADA was $0.26874086 (¥42.50), +9.84% from $0.24466542 (¥38.67) at the previous boundary. Because that is taken at a different time from the morning observation, it does not match the weekly +0.12%. ADA rose in the middle of the week and gave it back toward the end.
For NIGHT, the October 3 value is an exchange price quoted in USDT, while the October 10 value is derived from Cardano DEX prices; the source changed. A ratio of the two cannot separate price movement from the change of source, so we leave it out of the table. On a single source, CoinGecko’s seven-day series (as of 07:30 JST on October 10) shows NIGHT at −7.62%; ETH on the same series is −6.87%.
We do not connect price movements to the events covered in this brief. That includes the transaction surge in epoch 660 and ADA’s moves.
The rest of the network figures: epoch 659 had 21,207 blocks and 158,011 transactions, about the same as the epoch before. The staking ratio went from 56.18% to 56.16%, delegators fell by 553 to 1,334,182, and pools that produced blocks went from 952 to 933. The treasury stands at 1,374,649,329 ada in epoch 660.
2. Epoch 660: 1.08 Million Transactions in Three Days
An epoch is Cardano’s five-day unit. Recent epochs carried 130,442 transactions (657), 159,916 (658) and 158,011 (659), hovering around 150,000.
| Epoch | Period (JST) | Blocks | Transactions | Per block | Fees |
|---|---|---|---|---|---|
| 659 | 10/2 6:44 – 10/7 6:44 (5 days) | 21,207 | 158,011 | 7.45 | 49,494.70 ada |
| 660 | 10/7 6:44 – (3.02 days as of 10/10 7:16) | 12,897 | 1,083,871 | 84.0 | 205,533.61 ada |
Little more than three days in, epoch 660 had carried 6.86 times the transactions of the whole previous epoch. Scaled by elapsed time, that is a pace of about 1.79 million over five days. The rise has not been steady, though: from 922,251 at 14:30 JST on October 9, the count added only a little over 160,000 in the roughly 17 hours to the morning of October 10.
What they were: HOSKY’s “Doggie Bowl”
At 06:45 JST on October 7, about 30 seconds after epoch 660 began, the official account of HOSKY, a meme token from Cardano’s early days, posted:
“THE HOSKY DOGGIE BOWL IS NOW OPEN. We already hold the per epoch transaction record on Cardano for 5+ years (1.5M transaction). Now we’re coming back to break it. Send 2 ADA → get 1.XX ADA back + a random amount of $HOSKY.”
That afternoon it added: “We are trending to about 500k TX this epoch. We’re gonna need a lot more TX.”
We took three consecutive blocks from around 07:22 JST on October 10 and read the first 60 transactions of each, 180 in all, through Koios (a public API). 136 of them contained an output sending exactly 2 ADA to the same address (addr1vy740r73…tyns4). It is the same address SIPO found when it checked five other blocks on October 9.
In that October 9 check, transactions leaving this address sent about 1.86 ADA plus HOSKY tokens to around 20 recipients at a time. The token’s policy ID matched the one registered as “HOSKY Token” in Cardano’s token registry. Send 2 ADA, and HOSKY comes back together with the minimum ADA needed to receive it (Cardano transaction count reaches 5.8 times the previous epoch in just over two days, Japanese).
The announcement and the flow we observed on chain match. The announcement does not say how long the campaign will run or how the amount of HOSKY returned is determined, and on-chain records alone cannot confirm that the sampled address belongs to the HOSKY team. What we can write is that transactions of the announced kind, sending 2 ADA and getting HOSKY back, made up most of the count in the range we sampled. IOG, the Cardano Foundation and Intersect have not commented on the surge.
Blocks are close to full
The three sampled blocks held between 162 and 196 transactions, and their sizes ranged from 87,364 to 88,022 bytes. With a block size limit of 90,112 bytes, about 97% is in use. Because each transaction is small, well over 100 fit into a single block.
When blocks fill up, other transactions can be pushed to later blocks. How long ordinary transfers or DeFi transactions actually waited cannot be read from the records we checked. It is worth watching wallet and DEX announcements for any reports of delays.
Fees go into the reward pot
As of the morning of October 10, fees in epoch 660 were 205,533.61 ada, 4.15 times the 49,494.70 ada of the whole of epoch 659. Cardano fees are added to that epoch’s staking reward pot. Rewards paid to pools and delegators are fixed by the calculation after the epoch ends, so how much is added can be checked once epoch 660’s rewards are final.
SIPO’s three pools produced 110 blocks in epoch 659: 59 for SIPO, 29 for SIPO2 and 22 for SIPO3 (98 in epoch 658). From the block producer’s side, producing a block full of transactions is the same procedure. What differs is how many transactions are packed in, and the fees they carry.
3. The k Poll: Explicit No Reaches 26 Pools and 1 Billion ada
Should stakePoolTargetNum (k) be raised from 500 to 1000? (SPO poll) is now in its 23rd day since submission. The gap we described last issue — No ahead both by count and by stake — widened further in one week.
| Category | Last issue (10/3) | This issue (10/10) |
|---|---|---|
| SPO explicit Yes | 14 pools · 55,646,107 ada | 17 pools · 67,978,699 ada |
| SPO explicit No | 17 pools · 505,040,034 ada | 26 pools · 1,005,266,980 ada |
| SPO not voted | 8,737,025,845 ada | 8,544,163,377 ada |
| SPO always abstain (excluded) | 619 pools · 11,919,381,660 ada | 610 pools · 11,596,679,562 ada |
| DRep Yes | 23 votes · 123,625,124 ada | 29 votes · 137,633,650 ada |
| DRep cast No | 17 votes · 395,391,571 ada | 29 votes · 709,223,462 ada |
| DRep Abstain | 20 votes · 1,417,966,305 ada | 31 votes · 1,549,897,984 ada |
| Committee | 1 Yes · 1 Abstain | 1 Yes · 1 Abstain |
SPO votes added this week: 3 Yes and 9 No.
| Recorded (JST) | Pool | Vote | Delegated stake (approx.) |
|---|---|---|---|
| Oct 3, 11:03 | CHIL | Yes | about 11.15M ada |
| Oct 3, 16:32 | CROW | No | about 61.98M ada |
| Oct 3, 17:51 | GNP1 | Yes | about 0.94M ada |
| Oct 5, 23:16:33 | CAG · CF1 · CF2 | No | about 228.99M ada in total |
| Oct 5, 23:16:52 | CF3 · CF4 | No | about 96.55M ada in total |
| Oct 7, 01:03 | PSP | Yes | about 0.12M ada |
| Oct 7, 09:47–09:52 | SIPO · SIPO2 · SIPO3 | No | about 103.57M ada in total |
The five pools on the night of October 5 were recorded within 19 seconds. CF1–CF4 are pools whose registration data gives the name “Cardano Foundation – Private” and the Cardano Foundation’s website as the homepage. CAG is registered as “Cardano AG Pool – Private.” Together the five hold about 325.54 million ada, roughly two thirds of the No stake added this week.
All five votes point to the same rationale, authored by “Cardano Foundation.”
“The Cardano Foundation votes NO. In our view, raising stakePoolTargetNum (k) from 500 to 1000 would redistribute a fixed reward pot without addressing the factors that determine whether independent operators can remain viable.” “More pool IDs are not more operators.“
The rationale says this view also applies to the Foundation’s DRep vote, and adds: “This is a position on this specific proposal. It is not a vote against lower-saturated or community-run pools, or against decentralization.”
Distance to the threshold
The determination declared in the action’s text is whether explicit Yes exceeds 50% of active voting stake, with registered stake that does not vote counting the same as No. That denominator is now 9,667,470,771 ada, and the approval rate is 0.70%. Crossing 50% would take about 4.77 billion ada more in explicit Yes. About 8.54 billion ada has not voted, and the deadline is 06:44:51 JST on October 22.
No divided by Yes went from 9.08 times last issue to 14.79 times. This week Yes grew by three pools and about 12.33 million ada; No grew by nine pools and about 500.23 million ada. The DRep column moved too, but the action’s text declares that DRep votes are not counted in this poll’s determination.
4. SIPO’s Three Pools Voted No; Its DRep Abstained
This is SIPO’s own vote, so we disclose the record, the reasons and our interest together.
On October 7 (JST), SIPO voted No on the poll with its three pools — SIPO (09:47:09), SIPO2 (09:49:00) and SIPO3 (09:52:25) — and recorded an Abstain as a DRep at 10:06:11. The full rationale is published in Japanese and English at SIPO: No from three pools on the SPO poll to raise k to 1000.
The core reason is one point, and it looks at the same thing as the Foundation’s rationale in Chapter 3. The action’s own text says that raising k leaves “the total reward pot per epoch unchanged, other things being equal.” A change that keeps the total fixed and only re-divides it moves delegation between operators and does not grow Cardano. SIPO’s position is to oppose a raise that is detached from any design for enlarging what is shared. Not voting would have counted the same as No in the determination, but we chose an explicit No to put the position on record. Because DRep votes are not counted, SIPO abstained as a DRep.
The saturation line and SIPO’s three pools
This is a mechanical estimate: on-chain circulation of 36,797,328,639 ada on the morning of October 10, divided by k.
| k | Saturation point (circulation ÷ k) | SIPO’s three pools (live stake at end of epoch 659) |
|---|---|---|
| 500 (current) | about 73,594,657 ada | all three below |
| 1000 (proposed) | about 36,797,329 ada | SIPO 52,642,220 ada = above (1.431×) / SIPO2 34,812,587 ada (0.946) / SIPO3 24,570,855 ada (0.668) |
SIPO’s first pool sits above the k=1000 saturation point and would be disadvantaged if this change went through. SIPO2 grew in one week from 25,387,648 ada to 34,812,587 ada, about 9.42 million ada, and now stands at 94.6% of the k=1000 line. The reason for the increase cannot be identified from on-chain records.
SIPO could adapt to a lower saturation point by redistributing stake among its three pools. The rationale acknowledges this and states that community operators who built up delegation in a single pool have no such option. SIPO’s No should be read with its revenue impact in view.
Saturation caps rewards; it does not affect the right to produce blocks. Delegators can check today, without waiting for the result, which side of the line their pool is on.
5. minPoolCost: The Approval Rate Doubled in a Week
For Reduce minPoolCost to 75 ada, the last votes are accepted at 06:44:51 JST on October 12. It is an economic parameter change, and SPO votes are not required.
| Category | Last issue (10/3) | This issue (10/10) |
|---|---|---|
| DRep Yes | 17.42% (48 votes · 665,061,604 ada) | 34.45% (76 votes · 1,281,325,258 ada) |
| DRep cast No | 27 votes · 501,615,082 ada | 32 votes · 517,623,497 ada |
| DRep Abstain (excluded) | 12 votes · 1,373,729,188 ada | 17 votes · 1,477,728,444 ada |
| DRep not voted | 2,514,718,054 ada | 1,784,357,894 ada |
| Committee | 3 Yes · 4 not voted | 3 Yes · 4 not voted |
In one week, Yes grew by 28 votes and about 616.26 million ada. Unlike last issue’s “share of the rise that came from a shrinking denominator,” most of this week’s rise came from Yes votes themselves. The largest Yes votes, in the order recorded, are below. Voting power is each DRep’s current value as fetched on the morning of October 10.
| Recorded (JST) | DRep | Vote | Voting power (approx.) |
|---|---|---|---|
| Oct 3, 09:36 | CryptoCrow | Yes | about 286.36M ada |
| Oct 3, 10:49 | Rodrigo-[CHIL] | Yes | about 15.89M ada |
| Oct 4, 21:38 | wolf31o2 | Yes | about 24.04M ada |
| Oct 5, 19:07 | Everstake | Yes | about 192.37M ada |
| Oct 6, 05:17 | DraculaDAO | Yes | about 16.37M ada |
| Oct 9, 05:55 | Aichi/Tokai_DRep | Yes | about 30.83M ada |
| Oct 9, 10:46 | DINO | Yes | about 22.59M ada |
| Oct 10, 06:58 | JAZZ | Abstain | about 90.77M ada |
Reaching 67% takes 67.85% of those who have not voted
The denominator is now 3,719,406,264 ada. Reaching 67% needs 1,210,676,939 ada more in Yes. With 1,784,357,894 ada not yet voted, 67.85% of it would have to vote Yes before the deadline. More abstentions would shrink the denominator and slightly lower that figure. Votes can be changed until the deadline, and the committee needs five of seven Yes votes, so two more members.
What we can write is the arithmetic. Once the deadline passes, an action not ratified is recorded as “expired,” which is a different status from being rejected.
Disclosure. SIPO’s DRep abstained on this action on September 15 (SIPO DRep: Abstain on the standalone minPoolCost 75 ADA action). All three pools declare a fixed cost of 340 ada, so lowering the minimum to 75 ada has no mechanical effect on them.
6. OpenZeppelin Heads to Its Deadline Without a Path to 67%
Withdraw 11,787,063 ada for the OpenZeppelin Stack administered by Intersect closes at the same 06:44:51 JST on October 12.
| Category | Last issue (10/3) | This issue (10/10) |
|---|---|---|
| DRep Yes | 2.93% (10 votes · 129,704,259 ada) | 3.90% (15 votes · 169,525,021 ada) |
| DRep cast No | 53 votes · 1,431,202,398 ada | 82 votes · 2,024,470,932 ada |
| DRep Abstain (excluded) | 9 votes · 761,106,751 ada | 13 votes · 851,794,097 ada |
| DRep not voted | 2,733,110,520 ada | 2,015,245,043 ada |
| Committee | 3 Yes · 4 not voted | 3 Yes · 4 not voted |
The denominator is 4,345,340,611 ada. Even if all 2,015,245,043 ada not yet voted turned to Yes, approval would be 50.28%. Unless votes already cast change, it cannot reach the 67% threshold. Cast No is 11.94 times Yes, wider than 11.03 times last issue. Twenty-nine No votes were added this week, and some DReps, such as CryptoCrow and Everstake, voted Yes on minPoolCost and No on this action.
Disclosure. SIPO’s DRep is a party that voted No on this action on September 25. The rationale is at SIPO DRep: No on the 11.78M ADA withdrawal for the OpenZeppelin stack.
Treasury figures are unchanged from last issue. Withdrawals enacted since epoch 613 total 457,395,629 ada, leaving 42,604,371 ada under the 2026 limit of 500,000,000 ada.
7. CAP-12 and the Preparations for Dijkstra
Adding new parameters to the Constitution’s appendix
The constitutional amendment CAP-12 “Dijkstra era parameters” was submitted to Intersect’s CAP portal on October 2 and is in its comment period. On October 4, Intersect wrote on X:
“Dijkstra is the first Cardano hard fork that needs a constitutional change as well as a technical one. Without it, the new protocol parameters would be frozen at their launch values. CAP-12 adds them to the Constitution, with guardrails. Open for review now.”
The current Constitution says parameters not named in Appendix I may not be changed through governance. CAP-12 would add reference-script protection (4 items), Leios (9 items), Peras (6 items) and two pool-economics items (minimum pool margin and maximum pledge leverage) to the appendix, each with a permitted range. The 19 items for reference scripts, Leios and Peras would join the list of parameters whose change also requires SPO votes. Many of the upper and lower bounds are still undecided in the draft.
Disclosure. On October 3, SIPO commented on the two pool-economics items on the CAP portal, as a DRep that runs three pools (CAP-12 adds Dijkstra’s new parameters to the Constitution, Japanese). The comment asked that the minimum pool margin either be given bounds or be added by an amendment timed to the later hard fork that activates it, and that setting or lowering maximum pledge leverage come with an analysis of the stake that would fall outside rewards.
On October 9, CAP-12 came back as a revised draft. According to Intersect’s weekly update #132, the minimum pool margin, listed but never defined, now has its own section with guardrails. The minimum pool margin is meant to replace the minimum pool cost, and because an existing guardrail requiring the minimum pool cost to track the cost of running a pool would have blocked that very change, the author has moved to deprecate it. Remaining points, such as how often pledge leverage can change, were deferred to the following week, and the draft remains open. We could not find a stated comment deadline.
On the node side: 30% on older versions, and DijkstraNet next
The same #132 opens with this:
“block production on Cardano Mainnet for epoch 660 currently shows 30% still being validated by node versions older than 11.1.2”
11.1.2 was a critical update issued after a problem found on Preprod in September; Intersect recommends 11.1.3, with 11.1.2 as the minimum. SIPO’s three pools moved all their nodes to 11.1.2 by September 24.
For the next version, #132 says it will carry the full Dijkstra feature set minus Leios and launch a public network, DijkstraNet, for dApp and tooling integration. Version 11.3 is then planned to merge that with the Leios implementation into a hard fork release candidate. On the smart contract side, Essential Cardano’s weekly report of October 9 says Plutus 1.71.0.0 adds the V4 script context.
Peras adds a voting layer so that block finality can be recognized in as little as about two minutes. The initial implementation of its protocol parameters was merged into the ledger code at the end of September, but a change that bypasses Peras state handling in the next node release is also in. Updating the node alone will not switch Peras on. No mainnet activation date has been given.
Leios prototypes prototype-2026w40 (October 5) and 2026w40a (October 7) were released. The former resolves three audit findings (LEI-004, 005 and 009) and requires deleting the node database and resyncing to run.
8. Midnight Node 1.0.400 Discloses a Flaw Fixed First in Production
At 19:24 JST on October 6, Midnight’s node-1.0.400 was published on GitHub. It raises the ledger component to 8.1.3, fixing security advisory GHSA-wr7g-rr4v-jmj8. The advisory is rated 9.2 (Critical) under CVSS 4.0, an item raised in an audit report dated September 21.
The release notes describe this version as the public release of a security fix already deployed to three networks: preview, preprod and mainnet. It is a binary-only update that does not change the on-chain rules (the runtime), and no governance procedure is required.
Put simply, the advisory works like this. Midnight’s ledger verifies the zero-knowledge proof attached to a contract call as values in a world of remainders after division by a certain prime. The virtual machine that updates contract state, meanwhile, uses the original values as sent. So a value x and “x plus the prime” could be treated as the same value on the proof side and as different values on the VM side. The advisory gives the example of a contract that records paid invoice numbers to prevent double payment: resubmitting the same invoice in a different form could lead to repeated payouts. Ledger 8.1.3 stops accepting such non-canonical values inside contract calls.
Who needs to do what
Block-producing operators are asked to move to the same version together. An 8.1.3 node rejects some calls that an 8.1.2 node accepts, so mixed versions would split block producers’ judgments. Application developers are told they do not need to change how they build calls, because calls produced by the official compiler and midnight-js do not contain the rejected forms.
The advisory carries one caveat, though: rejecting these values from now on does not repair values already stored in contract state. Administrators of contracts that guard payouts or minting with sets of values such as invoice numbers are asked to check stored state for duplicates.
On timing: the advisory was published at 03:22 JST on October 2, and permissionless deployment on mainnet began at around 16:00 the same day. The release notes do not say exactly when the fix reached production. Neither the release notes nor the advisory says whether the flaw was ever exploited. This version also fixes the known 1.0.300 issue in which syncing mainnet from scratch stopped at block #1788979 (Midnight node 1.0.400 fixes a critical vulnerability, Japanese).
9. Midnight, the Week After Opening Deployment
In the week after deployment opened to everyone, moves to widen how Midnight is used sat alongside moves to get ahead of the parts that grow heavier with use.
- Three improvement proposals numbered: on October 5, MIP-0025, 0026 and 0027, along with MPS-0044, which defines a problem, were merged into the improvement proposal repository. MIP-0027 reuses zero-knowledge proof verification results, and reports cutting block production time by 42.4% and import time by 41.1% on a performance test network. Being numbered does not mean being adopted.
- TOKEN2049 keynote: on October 7, Midnight’s Fahmi Syed presented four products: Night Mode, Collateral Warehouse, Offer Files and Midnight Passport. Midnight’s official account summed them up as “the privacy layer for every chain.”
- Deployment requests retired: at 03:44 JST on October 10, a change ending the deployment-request process was merged into the improvement proposal repository. Stating that “Mainnet contract deployment is now permissionless,” it marks the request folder as a frozen archive and keeps the review rubric as optional self-assessment guidance. That answers one of last issue’s watch items.
- From Solana wallets: on October 7, Foundation CTO Sebastien Guillemot announced support for using Midnight from Solana wallets. It runs on testnet for now, and a mainnet release “is planned with the Midnight ledger v9 upgrade later this year.”
- Midnight City: Tokyo: held today, October 10, 15:00–19:00, at Happo-en in Shirokanedai, Tokyo. Charles Hoskinson and the Midnight City team appear; attendance requires the organizer’s approval.
Disclosure. SIPO is a member of Midnight’s Nightforce ambassadors.
10. CIP-113 Is Running on Mainnet
On October 7, at TOKEN2049 in Singapore, the Cardano Foundation announced that the programmable token standard CIP-113 was live on mainnet. The Foundation’s official account posted:
“Cardano’s programmable token standard, CIP-0113, is live on mainnet. Issuers of stablecoins and other regulated assets can now build compliance rules directly into native Cardano tokens. Enforced by the network itself. No hard fork required.”
A standard merged into the CIPs repository on September 30 reached production within a week. The same day, the Swiss industry body CMTA announced that it would treat CIP-113 tokens as equivalent to its tokenized-securities standard CMTAT. The wallets Eternl and GeroWallet, the explorer CardanoScan and the developer toolkit BloxBean, among others, supported it from launch. Intersect’s weekly update #132 says that the next day Veridian, the digital identity company spun out of the Foundation, tokenized its shares with CIP-113, making them the first asset under the standard. It also says the standard does not affect ada or existing native tokens.
As we wrote last issue, CIP-113 itself does not decide who can freeze or seize. That is set by each token’s rule components (modules). Which module a received token runs on becomes a new thing for holders to check.
11. DeFi
- Minswap and FluidTokens: on October 7, Cardano’s official account wrote in its TOKEN2049 day-one roundup that the two “announced their merger.” Meanwhile, asked “merge or collaboration?”, FluidTokens’ CEO replied that “Projects stay indipendent but teams are shared” (sic), and FluidTokens’ official account quoted that reply with “More details soon.” We have not confirmed a formal announcement by the two teams themselves. We do not call it a “merger” outright here and place the two descriptions side by side.
- SecondFi asset recovery tool: at 14:16 JST on October 7, SecondFi released a tool to recover assets from wallets affected by the June security incident. It is available only in a web browser on SecondFi’s official site, not inside the app. Processing can take up to five working days, and on October 8 SecondFi said the first batch of requests had been processed.
- RealFi on Lace mobile: at 05:31 JST on October 3, Lace temporarily paused RealFi on its mobile app while it works on improvements. It remains available on desktop. RealFi itself states its products are not available to users in the US, EU, UK, Hong Kong and other restricted jurisdictions. SIPO takes part in RealFi’s SPO accelerator as a RealFi alpha ambassador.
- FluidTokens deposits: after the September 30 pause caused by an oracle problem, FluidTokens announced on October 6 that “TVL is up 20% in the last 5 days.” The team itself says most of the growth comes from users chasing incentives.
12. Blockfrost Closes IceBreakers
At 23:11 JST on October 9, Blockfrost announced it is closing “IceBreakers,” the program in which SPOs and node operators served API responses and received a share of revenue.
“IceBreakers is closing. Operators can turn their nodes off. When we started, a community-run network serving real Cardano traffic was a theory. Over 100 operators proved it works, and works at scale.”
The announcement noted that higher throughput under Leios also raises the cost of indexing the entire chain, and said the funding model for the approach it had prepared — slicing the chain so operators share the work — was not approved by the community. Participating operators can turn their nodes off, and the amount and timing of final payments will be communicated individually.
The announcement does not name the proposal. The treasury withdrawal “Blockfrost: Maintenance and Next Generation Indexing,” which included that approach as one of its pillars, expired without ratification in epoch 633. Disclosure. SIPO voted Yes on that proposal as a DRep.
For SPOs, this ends one source of side income that could run next to a relay. The announcement recalls that Input Output invested in Blockfrost in January 2024 specifically to fund a move to a community-operated model, and that IceBreakers is what that turned into.
13. Organizations and People
- The Cardano Foundation announced, in quick succession, its September monthly report (October 6), the launch of the advanced CBCP certification exam (October 6), the spin-out of the digital identity business Veridian (October 8), and that its “.ada” top-level domain application had moved to the next phase of ICANN’s review (October 8). At the TOKEN2049 hackathon, CoinGraph, from the Cardano track, took the overall prize.
- Charles Hoskinson gave a main-stage keynote at TOKEN2049 on October 8; according to Intersect’s #132 he spoke about Midnight City and agentic payments and introduced a new concept, “Night Frames.” We do not quote the talk, as we could not check a transcript first-hand. On October 4 he listed the year’s work on X: “We just launched RealFI; Pogun is on the way; we launched Midnight; and now we have Midnight. city.”
- Teams building independent node implementations held a “Node Diversity Celebration Day” in Singapore on October 6. Dingo, a node written in Go, gained an interface in v0.78.0 that lets AI agents query the node’s database directly.
- In weekly update #131, Intersect reported that it is hiring two paid CIP editors, and that its own 2026/27 budget and Se7en Labs’ budget are now held in the dollar stablecoin USDM.
- Open Wallet Standard added Cardano support in v1.4.3. WingRiders built the implementation, with technical review from IOG and the Midnight Foundation.
- Midgard Labs, which took over Plutus development on October 1, holds meetups in Tokyo on October 23 and Osaka on October 24.
14. Risk Dimensions
- Network (transaction surge). In epoch 660, sampled blocks were filled to about 97% of the limit. How this affected waiting times for other transactions cannot be read from these records. HOSKY’s announcement does not say how long the campaign will run.
- Operations (Cardano nodes). According to Intersect, 30% of epoch 660 blocks are still produced on versions older than 11.1.2. The minimum is 11.1.2; the recommendation is 11.1.3.
- Governance (deadline).
minPoolCostand OpenZeppelin close at 06:44:51 JST on October 12.minPoolCostneeds Yes from 67.85% of non-voters and two more committee members; OpenZeppelin cannot reach 67% unless votes already cast change. - Pool economics (SIPO itself). SIPO’s first pool sits at 1.431 times the k=1000 saturation point, and SIPO2 has reached 94.6% of the line. SIPO is a party that voted No with three pools.
- Operations (Midnight node). The switch to 1.0.400 assumes block producers move together. Contract state already stored is not repaired by the new rule.
- Standards (CIP-113). Even among CIP-113 tokens, freezing and seizure powers are set by modules. The standard’s name alone does not tell you how a token behaves.
- DeFi (the word “merger”). For Minswap and FluidTokens, “merger” and “independent projects sharing a team” stand side by side. Whether deposited assets or contracts move will depend on a concrete announcement.
- Infrastructure (data provision). With IceBreakers closing, the community-run API network stops. Who bears the cost of indexing after Leios remains open.
- Constitution (Dijkstra). Intersect explains that if Dijkstra launches before CAP-12 is in place, the new parameters will be frozen at their launch values.
15. What to Watch Next Week
- The
minPoolCostand OpenZeppelin deadline — 06:44:51 JST, October 12. WhetherminPoolCostYes can capture 67.85% of non-voters, and whether a fourth and fifth committee vote arrive. Check each action’s page on gov.tools. - Epoch 660’s final transaction count — fixed at the same 06:44:51 JST on October 12. Check the epoch 660 pages on AdaStat and Cexplorer.
- Epoch 661’s count — whether it returns to around 150,000 or stays high. Check on the same two explorers at the boundary at 06:44:51 JST on October 17.
- Epoch 660 rewards — how the extra fees show up in rewards. Epoch 660 rewards are finalized after the boundary two epochs later (October 17). Check Koios epoch information.
- SPO votes on the k poll — which way the roughly 8.54 billion ada not yet voted moves. The deadline is 06:44:51 JST on October 22; check the Info Action’s page on gov.tools.
- Details from Minswap and FluidTokens — FluidTokens’ official account says “more details soon.” Watch for user-facing changes and migration steps on both teams’ official X accounts (@minswap · @FluidTokens).
- Midnight’s advisory page — whether results of checks on existing contracts are added to GHSA-wr7g-rr4v-jmj8. Check the advisory page on
midnightntwrk‘s GitHub. - The next node version and CAP-12 — the next version carrying the full Dijkstra feature set minus Leios and the launch of DijkstraNet, plus the remaining CAP-12 points (how often pledge leverage can change, and its relation to CAP-10) that the author says will follow next week. Check the
IntersectMBO/cardano-nodereleases page, Intersect’s next weekly update (#133), and CAP-12’s page on the CAP portal. - Cardano PRIME’s call deadline — 12:59 JST on October 15. Check Cardano PRIME’s announcements.
- Midgard Labs meetups — October 23 (Tokyo) and 24 (Osaka). Check the announcement on X by Midgard Labs COO Dave Dionisio.
Conclusion
The biggest movement on Cardano this week was in the number of transactions. Epoch 660 carried 1.08 million in three days, 6.86 times the whole previous epoch, and sampled blocks were filled to about 97% of the limit. Most were the transactions HOSKY announced as the epoch began: send 2 ADA, get HOSKY back. HOSKY itself says it came back to break the per-epoch transaction record it set more than five years ago; whether it does will be known on the morning of October 12. What is certain is that fees piled up to 4.15 times the previous epoch and go into the reward pot.
Governance is heading toward an early-week deadline. minPoolCost‘s approval rate doubled in a week to 34.45%. This week’s rise came from Yes votes themselves, and still, reaching 67% would take a little over two thirds of those who have not voted. OpenZeppelin would reach 50.28% even if every non-voter voted Yes.
In the k poll, explicit No passed 1 billion ada. The Cardano Foundation voted No with five pools and wrote, “More pool IDs are not more operators.” SIPO’s three pools joined them, putting on record SIPO’s reason: it opposes a change that re-divides rewards without changing their total. It also disclosed that its first pool sits above the new line.
On Midnight, node 1.0.400 revealed that the fix for an advisory published in the early hours of the day deployment opened was already running in production. The reviewed deployment-request process was formally closed, and the review rubric became optional self-assessment guidance. In the week after anyone could deploy, the checks extend not only to new calls but to state already stored.
References
Where the text refers to SIPO’s own articles, links are given in place. Primary sources:
- Koios REST API (epoch information, block and transaction contents, action status, vote timestamps, DRep and pool voting power, treasury and circulation): https://api.koios.rest/api/v1
- Cardano governance actions: https://gov.tools/governance_actions
- k SPO poll (voting status): https://gov.tools/governance_actions/f6fd3678f12edc58dd8739560149fcdb0b8b6fc77a5f303d49c87c74d1fccb4c#0
- k SPO poll, action text: https://gateway.pinata.cloud/ipfs/bafkreia3u74dgyvg5v3u7wxwip44wwqzyhynqqm4wzyhkkwevaiynlihnm
Reduce minPoolCost to 75 ada: https://gov.tools/governance_actions/75e7882a8ef2bc39517bffbfb654e89f525def5a8364d81e11fc5facafc6dd9b#0- OpenZeppelin withdrawal action: https://gov.tools/governance_actions/418df5986f50547ec4a709f1a5bce6b850753ad6614c873386b7c953fee84a9f#0
- HOSKY’s Doggie Bowl announcement (October 7): https://x.com/hoskytoken/status/2107588050771722598
- HOSKY’s follow-up (October 7): https://x.com/hoskytoken/status/2107740787375313218
- Cardano Foundation rationale (k poll, five pools): https://ipfs.blockfrost.dev/ipfs/QmScoc5qaitVUfnwmHGMEvPXhWLb42HYdpXCWNVent6V57
- Intersect weekly update #132 (node version distribution, CAP-12 revision, DijkstraNet, CIP-113 and Veridian): https://intersectmbo.org/news/intersect-weekly-update-132-oct-9-2026
- Intersect’s CAP-12 announcement (October 4): https://x.com/IntersectMBO/status/2106702381853364701
- Essential Cardano weekly development report (2026-10-09): https://www.essentialcardano.io/development-update/weekly-development-report-2026-10-09
- Leios prototype: https://github.com/input-output-hk/ouroboros-leios/releases/tag/prototype-2026w40
- midnight-node node-1.0.400: https://github.com/midnightntwrk/midnight-node/releases/tag/node-1.0.400
- Midnight security advisory GHSA-wr7g-rr4v-jmj8: https://github.com/midnightntwrk/midnight-ledger/security/advisories/GHSA-wr7g-rr4v-jmj8
- MIP-0027: https://github.com/midnightntwrk/midnight-improvement-proposals/blob/main/mips/mip-0027-transaction-revalidation-cache.md
- Retirement of the deployment-request process (improvement proposal repository #375): https://github.com/midnightntwrk/midnight-improvement-proposals/pull/375
- Midnight’s official TOKEN2049 post: https://x.com/MidnightNtwrk/status/2107683779447816316
- Sebastien Guillemot’s post (Solana wallet support): https://x.com/SebastienGllmt/status/2107526584928555372
- Midnight City: Tokyo announcement: https://x.com/MidnightCitySim/status/2106670673565532431
- Cardano Foundation’s CIP-113 launch announcement: https://x.com/Cardano_CF/status/2107667227210367411
- Cardano’s official TOKEN2049 day-one roundup: https://x.com/Cardano/status/2107872852015739139
- FluidTokens CEO’s reply: https://x.com/0xMetamatt/status/2107747783696371910
- FluidTokens’ “More details soon”: https://x.com/FluidTokens/status/2107759346889236914
- SecondFi asset recovery tool announcement: https://x.com/secondfiapp/status/2107701581160517828
- SecondFi first batch processed: https://x.com/secondfiapp/status/2108039078180733383
- Lace’s RealFi pause announcement: https://x.com/lace_io/status/2106119985135284328
- Blockfrost’s IceBreakers closing announcement: https://x.com/blockfrost_io/status/2108560910025375996
- Cardano Foundation (September activities): https://x.com/Cardano_CF/status/2107435961185366161
- Cardano Foundation (CBCP): https://x.com/Cardano_CF/status/2107506978830356968
- Cardano Foundation (Veridian): https://x.com/Cardano_CF/status/2108044721410953268
- Cardano Foundation (.ada): https://x.com/Cardano_CF/status/2108088602852786427
- Cardano Foundation (hackathon): https://x.com/Cardano_CF/status/2108218814160662878
- Hoskinson’s post (October 4): https://x.com/IOHK_Charles/status/2106571628482121772
- Node Diversity Celebration Day announcement: https://x.com/IntersectMBO/status/2107093675511517341
- Open Wallet Standard Cardano support (IOG): https://x.com/IOGroup/status/2108118456046690603
- Morning fixed values and seven-day series (ADA / BTC / ETH / NIGHT): https://www.coingecko.com
- Weekly Brief #26: https://sipo.tokyo/language/en/weekly-brief-26-en/
Transparency Notes
- Epoch 660 count and ratios: values for the in-progress epoch fetched from Koios at 07:16:20 JST on October 10. 6.86× is 1,083,871 ÷ 158,011; per-block figures are 1,083,871 ÷ 12,897 and 158,011 ÷ 21,207; about 1.79 million is a mechanical estimate dividing 1,083,871 by 3.02 elapsed days and multiplying by 5. The 4.15× fee figure is 205,533.61 ÷ 49,494.70.
- Share of 2 ADA transactions: a sample reading only the first 60 transactions of each of three consecutive blocks (14,047,733–14,047,735) around 07:22 JST on October 10. 136/180 is not a share of whole blocks or of the epoch. The ~97% block fill is the three blocks’ sizes divided by the 90,112-byte limit.
- HOSKY being sent back: the campaign’s description comes from HOSKY’s official announcement (06:45 JST, October 7). The return transactions were checked on chain in SIPO’s October 9 reading of five blocks and in a separate two-block check; this issue’s three-block sample did not read the outgoing transactions. We have not confirmed that the sampled address belongs to the HOSKY team. The past record HOSKY claims (1.5 million) is HOSKY’s own statement and was not verified here.
- k poll denominator and approval rate: Yes of 67,978,699 ada plus the stake counting as No (explicit No, always no confidence, not voted) of 9,599,492,072 ada gives a denominator of 9,667,470,771 ada and an approval rate of 0.703%. Explicit abstentions and the 610 “always abstain” pools are excluded, as the action’s text declares.
- Per-pool stake: per-pool values in Chapter 3 are each pool’s active stake for the current epoch fetched on the morning of October 10, a different route from the action’s aggregated totals. They are therefore given as approximations, and totals use the aggregated figures.
- The Cardano Foundation’s five pools: the registered names of CF1–CF4, and the fact that all five votes point to the same rationale (author: “Cardano Foundation”), were checked against on-chain records and the rationale text. We do not describe CAG’s operator beyond the rationale’s reference to “the Foundation’s stake pools.” We have not checked when the Foundation’s DRep vote, to which the rationale also applies, was recorded.
- DRep voting power: values in the Chapter 5 table are each DRep’s current value fetched on the morning of October 10 and may differ from the value at the time the vote was recorded.
- minPoolCost’s 67.85%: 0.67 × 3,719,406,264 − 1,281,325,258 = 1,210,676,939 ada, divided by the 1,784,357,894 ada not voted (the stake counting as No minus cast No and always no confidence of 136,099,615 ada). Changes in abstentions are not included.
- OpenZeppelin’s 50.28%: Yes of 169,525,021 ada plus 2,015,245,043 ada not voted, divided by the denominator of 4,345,340,611 ada.
- Why we do not write “X% against”: the No rate returned by the public API includes non-voters and “always no confidence.” We therefore give cast No only as vote counts and ada amounts.
- Saturation estimate: on-chain circulation of 36,797,328,639 ada on the morning of October 10, divided by k. SIPO’s three pool values are live stake fetched right after the end of epoch 659. The reason for SIPO2’s increase cannot be identified from the records, so it is not stated.
- SIPO’s interests: for the k poll, that SIPO is a party that voted No with three pools and abstained as a DRep and that its first pool sits above the k=1000 saturation point; for
minPoolCost, that all three pools declare a fixed cost of 340 ada and see no mechanical effect and that SIPO’s DRep abstained; for OpenZeppelin, that SIPO’s DRep voted No; for Blockfrost, that SIPO voted Yes as a DRep on the expired treasury proposal; for CAP-12, that SIPO commented on October 3; for RealFi, that SIPO takes part in the SPO accelerator as an alpha ambassador; and for Midnight, that SIPO is a member of the Nightforce ambassadors. - Times of market values: the October 3 values are from 06:35:29 JST and the October 10 values from 06:35:12 JST, from our fixed morning observation. ADA on October 10 was recorded to two decimal places ($0.24), and the weekly +0.12% is computed from that value. The epoch-boundary ADA (+9.84%) is taken exactly at the boundary and at a different time from the morning value.
- NIGHT values: $0.04617263 on October 3 is an exchange price quoted in USDT; $0.045657 on October 10 is derived from Cardano DEX prices. Because the sources differ, the table leaves the weekly change blank. The −7.62% for NIGHT and −6.87% for ETH in the text come from CoinGecko’s seven-day daily series (09:00 JST October 3 → 07:29 JST October 10), a different series and time from the morning observation.
- The word “security” for Midnight 1.0.400: the release notes say “security patch” and “critical advisory,” and the advisory’s severity is “critical.” We use the terms within that range. Neither the release notes nor the advisory uses the word “vulnerability” or gives a CVE number.
- Minswap and FluidTokens: we place Cardano’s official account’s “merger” alongside the FluidTokens CEO’s “projects stay independent but teams are shared,” without leaning toward either. We have not confirmed an announcement from Minswap.
- What we did not write: Hoskinson’s “within 90 days” remark about Pogun (not found in his posts this week), Pulse’s token-launch mechanism on Midnight (no first-hand announcement confirmed), what Night Mode and the other products do (only names published), a CAP-12 comment deadline (none stated), and a DRep’s claims about the drafting of the k poll’s text (facts not verifiable).
- Veridian and CIP-113: that Veridian’s shares became the first CIP-113 asset is based on Intersect’s weekly update #132.
- Prices and events: not connected.
This is not investment advice. For information and research purposes only.