Version: 1.0
Status: Reference draft
Date: 2026-10-18

Purpose

These notes provide background and review guidance for MSIG #5, including the launch checklist.

Reference only — not an additional resolution or an exhibit approved by MSIG #5.

These notes preserve reasoning from MSIG #5 v6.14, ingested on 6 October 2026. That is an ingestion date, not a verified market-observation date. Market figures, elapsed-time descriptions, and claims about prior review are historical and have not been refreshed or independently reverified here. Operative requirements remain in MSIG #5 and its Framework, including the annual pay review and treatment of existing Treasury contracts.

Notes for review

1. What the vstcreation MSIG says. The full text has now been reviewed. Four points.

No Committee term exists. vstcreation establishes the Oversight Committee "with three (3) initial members ... effective upon formation of the Trust" and stops there. No duration, no renewal, no selection process for later appointments. Framework 3.2–3.4 now sets ordinary and staggered initial terms; MSIG #5 Part B ends the former members’ transition service when all five seats are filled. MSIG #2’s reference to "the remainder of Dario Cesaro’s Initial Term" did not itself specify a duration.

The likely source of the confusion. The Trust was formed on 13 February 2026. The Committee has been serving for roughly six months. That is an elapsed period, not a term — and it is easy to hear one as the other. The only six-month term anywhere is the Trustee's Initial Interim Term under Trust Agreement section 5(b).

Two approved documents remain unreviewed. vstcreation approved the LLC Operating Agreement, the Trust Agreement, and the Trustee Compensation and Indemnification Acknowledgment. Only the Trust Agreement has been provided. The standard Independent Contractor Agreement has since been reviewed and Part H now follows it, but the Acknowledgment may add indemnification terms specific to compensated governance roles that Committee members should have too. Worth checking before Part H is final.

The amendment path is clear. vstcreation provides that material amendment of the approved governing documents requires a separate MSIG Resolution. Part J relies on that clause.

2. Where the figures come from. The operative amounts are proposed in MSIG #5, not approved by these notes. The source draft said the working group had confirmed them. Framework 13.1 permits changes at the Threshold; the reasoning below is retained as historical context. The market figures are an undated historical illustration, not current values; USD limits are unaffected by price movement, but the share-of-yield comparison is not. A trades at roughly USD 0.077, giving a market capitalisation of about USD 127 million on a circulating supply of about 1.66 billion A. The REX yield pool of 18–20M A a year is therefore worth roughly USD 1.4–1.5 million a year, or about USD 350,000–380,000 a quarter.

3. What this costs against the yield, over a year. Quarterly figures make the total easy to miss, so here it is plainly.

The gross figure is the one that leaves the Treasury. The 25% margin is funded once, not four times — each instalment after the first tops the account up to 125% of the coming cycle's spend, so it restores whatever the previous cycle consumed rather than adding a fresh margin. At full spend and a steady price the gross is 284,375 + 3 × 227,500. The margin is not spending authority and is returned at the end of the funding period, but it is unavailable to the Network in the meantime.

Gross runs higher than 966,875 if the price of A falls, because the top-up restores a USD-denominated level from an account holding A. It runs lower if the program underspends, since a carried balance reduces the next top-up.

That is a large share, and block producers should decide it deliberately. The REX yield is not reserved for the RFP program — it also has to cover Labs, infrastructure, and anything else the Network funds from it. If the intention is for the RFP program to be one call on the yield among several, the quarterly ceiling should come down. A ceiling of USD 100,000 a cycle brings the annual Total Program Spend to USD 710,000 — 400,000 in awards plus 310,000 of Program Costs — about 49% of the yield.

The recommendation of USD 150,000 a cycle stands on administrative grounds — five part-time people can run perhaps four to eight RFPs well in a quarter, at USD 15,000 to 30,000 each, plus the in-cycle portion of any recurring infrastructure. But that is a capacity argument, not an affordability argument, and the original draft recommended reconciling them before a vote. This commentary adds no approval condition.

4. What the Delphi Oracle actually provides. The contract was reviewed rather than assumed. Three things matter.

It publishes a median of the last 21 oracle submissions, held in the median field of a datapoints table scoped by pair. That is a near-real-time figure, not a time-weighted average. An earlier draft specified a 30-day moving average; the contract does not provide one. The bars table in the header, which would hold aggregates, is declared but never populated and has no table type — it is unused code.

It keeps no price history. Only 21 rows exist per pair and the oldest is overwritten on each submission, so a rate read today cannot be re-read from the table tomorrow. That is why the approval record must capture the block and transaction of the read.

The smoothing does not matter as much as it first appeared. Because awards are denominated in USD, an awardee receives their contracted USD value whatever the price of A is doing — a volatile rate does not change what they are paid. The volatility lands on the program's A outflow instead, and that is already managed by the coverage test and the 25% margin. The 15% collar in Framework 13.4 covers the remaining case: a momentary spike or crash producing an anomalous payment.

4b. Denomination, and who carries the price risk. An earlier draft priced awards in A and froze the A amount at the award decision, which put the price risk on awardees. That has been reversed: awards are now denominated in USD and the A amount is calculated at each milestone approval, so the program carries the risk.

This is the better allocation. Awardees are teams and individuals who budget in fiat, and asking them to absorb the movement of A over a multi-month engagement would have shown up as padded bids, shorter engagements, or good teams declining to bid at all.

The cost is that the program's purchasing power now moves with A. That is what the coverage test in Framework 13.4 and the 25% funding margin in Part I exist to manage. Two consequences block producers should understand before voting:

• A sustained fall in A can force the program to stop awarding mid-cycle. The coverage rules say so plainly rather than leaving it to be discovered.
• The Treasury may be asked for a top-up if coverage falls to the 10% floor. That is a reserved matter and comes back to block producers.

5. How a USD 100,000 award sits inside a USD 150,000 cycle. Charged to one cycle it would consume two-thirds of it, leaving little for anything else — which is why multi-cycle awards are handled separately in Framework 13.2. A twelve-month infrastructure service at USD 7,000 a month reserves only the months falling inside the current cycle, around USD 21,000 for a quarter. The rest becomes a forward commitment that reduces the ceiling in each later cycle it touches.

Two guardrails come with that. Forward commitments are published in every cycle report, so the Network can see what future budgets are already spoken for. And an award extending past the authorized funding period is a reserved matter, so the Committee cannot commit the Network beyond the funding block producers have approved.

There is also a practical point about the existing arrangement. The Treasury already contracts these services directly. Bringing that into the RFP program is a separate decision — whether to re-compete the service through an RFP, or to move the existing contract across as it stands. Neither happens automatically on approval of this MSIG, and whichever is chosen should be done deliberately rather than by the Committee simply issuing an RFP over a live arrangement.

6. Why the mandate has no end date, and what replaces one. The mandate runs until block producers cancel it, which is what a standing body normally looks like and avoids a cliff where the program stops because a renewal vote was not organized in time. The funding period, by contrast, does expire — every four cycles — and that is where the recurring decision now sits.

An expiry date was doing one useful thing, though, and it should be replaced rather than dropped. It forced a periodic decision. Without it, inertia favours continuation: revoking takes 15 of 21 affirmative votes, so a Committee that is merely mediocre rather than failing will continue by default.

Two things take its place:

• Funding is authorized a year at a time. Four cycles per Resolution, transferred quarterly. The Committee can only ever spend what block producers have authorized, and declining to authorize the next year requires no confrontation and no revocation vote. Block producers can also stop or reduce an instalment mid-year at the Threshold.
• An annual published review. Awards, outcomes, conflicts, coverage history, and whether the limits still fit. A scheduled moment when performance is on the record, whether or not anyone calls a vote.

This also fixes the awkward interaction with long awards. The constraint on a multi-cycle award is no longer "does the mandate still exist" but "is it inside the authorized funding period" — which is a better test, because it points at money block producers have actually committed rather than at the Committee's own tenure.

7. The standard contractor agreement covers more than expected. VS LLC's Independent Contractor Agreement template — the instrument already executed for the Trustee and LLC Manager — is used for Committee members, Program Managers, Technical Reviewers, and awardees alike, with a role-specific Schedule A. Four of its provisions do work this MSIG would otherwise have had to duplicate:

• Section 2 provides that the agreement does not create a governance role. The seat comes from the MSIG appointment; the contract covers services, pay, confidentiality, and IP. That is exactly the right separation for a Committee member.
• Section 5 terminates the services automatically when a role requiring MSIG appointment is ended by MSIG Resolution. Removal and contract termination stay in step without further drafting.
• Section 9 vests work product in the Trust regardless of funding source, in terms almost identical to the Trust Agreement. An earlier draft of this MSIG assigned work product to VS LLC; that was wrong and has been corrected to match the executed template.
• Section 11 requires return or secure deletion of credentials on termination, and cooperation in credential rotation and revocation of permissions. That is the contractual counterpart of the key-surrender obligation in the Framework.

8. Why the clocks start at funding. Under MSIG #3 the Treasury funding for the VST was approved well before it arrived, and the Trustee and LLC Manager contracts were only signed in June 2026 once it did. If the same gap happens here and the cycle and pay run from MSIG approval, VS LLC would carry retainer obligations against money it does not have, and the cycle would burn while the Committee had nothing to allocate. Framework 13.1 ties both clocks to the arrival of funds.

9. Settled for now: equal pay. Part H pays all five seats equally, and the working group has confirmed that for the first funding period. Revisit at the annual review under Framework 14.3 — and note that Reviewer fees under Framework 12.6 make pay unequal in practice. The working group has decided not to cap that inequality by amount: the concurrent-engagement cap in Part H bounds the workload, and the cycle report makes the resulting totals visible. Unequal pay is therefore accepted and published, not capped. The arguments considered, kept as the record of reasoning:

For differentiating. Core Development, Business Development, and Marketing carry recurring scoping and diligence work in their categories. Technical judgment in particular is the hardest of the five to recruit, and a weak evaluation there costs the most.

For equal pay. By the seat table in the Framework, the Community seat carries two named MSIG #4 categories — educational initiatives, and community engagement and advocacy — while Business Development carries one. Portfolio load does not rank the way intuition suggests, and it is in any case a poor proxy for effort. The At-Large member chairs the Committee, which is different work rather than less of it. All five hold identical vote weight, identical signing duty, identical key custody, identical conflict obligations, and identical liability exposure. And the Community seat is the one selected through community vote — paying it least is a statement about its standing that will be quoted back during the first contested award.

Alternatives to a seat differential. Scarce technical expertise can be bought per engagement through the Technical Reviewer budget, where it is needed and at a rate that reflects it, rather than embedded permanently in a retainer. Drafting and category diligence can sit with the Program Manager pool, with the portfolio lead sponsoring and directing rather than producing.

10. Read Part I narrowly. The Treasury has said the open question is who represents the Network to receive and manage these funds. The Steering Committee serves that purpose for the RFP program only. Framework 8.4 requires a separate MSIG Resolution before the Committee, VST, or VS LLC may form or join an entity on behalf of the Network.

Drafted to match the structure of MSIGs #2 to #4.

Framework launch recommendation

Recommendation: do not run the open call in cycle 1 (Framework section 26). This is non-binding advice, not a program rule.

Launch checklist

This checklist tracks remaining launch tasks. It adds no rules or approvals; the blocking conditions below are established in MSIG #5, the Framework, and Exhibit D. The initial Manager pool requirement is in Framework 12.2.

Scope. This checklist also tracks off-chain candidate sourcing, counsel’s conforming amendments and operating-procedure changes, and the EOS Rio code and rights handover. It replaces references to the separate, unprovided Program Administration Register; these notes are not part of the Framework or approved by block producers. The rules cited below remain controlling.

Task numbers remain fixed. Delete resolved tasks without renumbering the rest.

Revision scope and remaining review

These are proposed replacement drafts for VST review, not versions already issued or approved by VST. The root proposal’s revisions record the changes; earlier comments and citations refer to their original snapshots and are not retroactively updated. VST’s upstream review and issuance remain outstanding before these can be treated as revised upstream source documents.

• Intentional: all five seats require future MSIG appointments; the three former members perform unpaid Trust oversight only during transition. The Community term starts at appointment and can run before program funding; the initial ECF route has the nine-month fallback in Framework 4.6.
• Intentional Service-IP policy: providers retain pre-existing IP; licences and handover materials are expressly agreed per RFP and contract, with switching limits published. The automatic embedded-IP licence back applies only to Deliverable and Embedded awards. Network data export, agreed handover rights, and the counsel gate remain.
• Contract cancellation: no-liability termination applies to the initial award disbursement under Framework 10.5 or 10.8, not later accepted work; freezes and mandate suspension/revocation preserve signed-agreement obligations under 10.9, 11.10a, and 14.4–14.5.
• Safeguards restored or clarified: ECF’s own funding MSIG, secure-key eligibility, immediate loss of signing weight on resignation notice, the vision-statement onboarding gate, and Committee-vote suspension of role/signing weight. Automatic contractual pay suspension and Treasury-affiliation suspension remain separate.
• Disclosure and recovery: VQ2 distinguishes public identity from private legal identification without reinterpreting VQ1 records; the seven-day oracle-recovery rule and resumption of suspended quarterly instalments remain explicit.
• Technical review: the proposer/canceler design follows the Antelope reference MSIG contract. This is a design reference, not verification of Vaulta’s deployed contract; Exhibit D requires deployed-code checks and testnet rehearsals before mainnet setup and funding.