KVORUM
compoundSource: Compound governor bravo · #376

Compound Sandbox: Development Proposal

executedDec 5, 2024, 07:02 AM UTCby3F0x3fb1…4c8a

01Summary

AISummary by Kvorumclaude-haiku-4-5

Compound DAO proposes to pay WOOF! team $600K USDC to develop Compound Sandbox, a permissionless money-market platform based on Compound v3. Payment structure: $300K upfront and $300K streamed over 6 months at 0.0193 USDC/sec. Development begins January 1 with testnet deployment targeted within 24 weeks.

Type
treasury allocation
Funding
$600000

Key changes

  • $300K USDC withdrawn immediately from cUSDCv3 via Compound Aera vault to WOOF! team(high)
  • $300K USDC streamed over 6 months (starting at proposal execution) via Llama Pay Router at rate of 0.0193 USDC/sec(high)
  • WOOF! team to develop and deploy Compound Sandbox testnet within 24 weeks from January 1 start date(high)

Notable concerns

  • Production deployment date depends on community feedback and audit timeline
  • Services governed by Terms of Service updated November 29, 2024
Contracts0xc3d688B66703497DAA19211EEdff47f25384cdc30x8624f61cc6e5a86790e173712afdd480fa8b73ba0x3b109aa111BdF11B30350CdfAd7e9Cf091421Aa40xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48
Generated by Kvorum
model: claude-haiku-4-5
provider: anthropic
profile: anthropic-fast-v1
prompt: v1.0
generated: 2mo ago
Read the description ↗

02Description

231 words

Compound Sandbox: Development Proposal

Background

WOOF! team offers to build Compound Sandbox - a Compound v3-based permissionless money-market platform, an extension of Compound Finance.

The scope of work is detailed in the forum post.

WOOF! team will start the development on January 1. WOOF! team will develop and deploy a testnet within 24 weeks. Production deployment date will vary on community feedback and audit timeline.

This proposal initiates a $600K payment in USDC: 50% will be available to WOOF! immediately, while the remaining 50% will be streamed over six months starting at the date of proposal execution.

Compensation Structure

With the adoption of Proposal 249, there is now a standard payment process for all Compound DAO service vendors utilizing payment streams in USDC.

The first proposal action calls withdrawTo on a cUSDCv3, via the Compound Aera vault’s execute function. The withdrawTo transfers upfront 300K USDC to WOOF.

The second proposal action calls createStream on a Llama Pay Router, via the Compound Aera vault's execute function. The stream will pay out on a per-second basis over 6 months. $300K USDC over 15552000 seconds gives a rate of 0.0192901235 USDC/sec.

WOOF! will be able to withdraw the vested funds at any time.

By approving this proposal, you agree that any services provided by WOOF! shall be governed by the Terms of Service that were updated as of November 29, 2024.

03Mismatch analysis

AIMismatch analysis by Kvorumclaude-sonnet-5
Consistent

I compared the two decoded execute() calls against the two actions described in the proposal text. Action 0 targets a Comet-like contract with a withdrawTo-style selector and an amount that decodes to 300,000,000,000 in 6-decimal USDC units (300K USDC), matching the described immediate 300K USDC payment. The recipient address (0xd36025e1e77069aa991dc24f0e6287b4a35c89ad) and USDC token address are consistent between both actions, supporting the claim that both payments go to the same party (WOOF). Action 1 targets a LlamaPay Router with a createStream selector; the duration parameter decodes exactly to 15,552,000 seconds, matching the '6 months' streaming period stated in the description, and the token/recipient match action 0. No numeric, target, or party mismatches were found between the prose and the calldata. The description also correctly attributes both calls as routed through the Aera vault's execute() function. Given the precise numeric match on both the upfront amount and the stream duration, and consistent addresses across both actions, the proposal description accurately and completely reflects the on-chain actions.The compensation structure section (referencing Proposal 249) is contextual and does not conflict with the calldata. Overall, this is a well-matched, low-risk proposal.due to exact match of key figures (300K USDC, 15,552,000 seconds) and consistent addresses across the two on-chain actions with the prose description.

Generated by Kvorum
model: claude-sonnet-5
provider: anthropic
profile: anthropic-strong-v1
prompt: v1.0
generated: 2mo ago
Review the decoded actions ↗

04Actions

2 total
#0execute((address,uint256,bytes))chain 0x1
Target
0x8624…73ba
Value
0 wei
Function
execute((address,uint256,bytes))
Arguments
{
  "0": [
    "0xc3d688B66703497DAA19211EEdff47f25384cdc3",
    "0",
    "0xc3b35a7e000000000000000000000000d36025e1e77069aa991dc24f0e6287b4a35c89ad000000000000000000000000a0b86991c6218b36c1d19d4a2e9eb0ce3606eb4800000000000000000000000000000000000000000000000000000045d964b800"
  ]
}
Calldata
#1execute((address,uint256,bytes))chain 0x1
Target
0x8624…73ba
Value
0 wei
Function
execute((address,uint256,bytes))
Arguments
{
  "0": [
    "0x3b109aa111BdF11B30350CdfAd7e9Cf091421Aa4",
    "0",
    "0x122f79c0000000000000000000000000a0b86991c6218b36c1d19d4a2e9eb0ce3606eb48000000000000000000000000d36025e1e77069aa991dc24f0e6287b4a35c89ad0000000000000000000000000000000000000000000000001ac53a820f8a61f90000000000000000000000000000000000000000000000000000000000ed4e00"
  ]
}
Calldata

05Tally

Summed from votes

Voters20
VP participating846.3K
Outcome (current)For
Current support100%

06Voters

20 total

No votes recorded yet.

07Forum

AIForum synthesis by Kvorumclaude-sonnet-5
sentiment · mixedthread · constructive

Arguments for

  • Sandbox provides permissionless infrastructure that could significantly grow Compound's TVL, treasury (via COMP buybacks), and strengthen the COMP token through fee-to-COMP conversion mechanisms.— Pasha, dmitriywoofsoftware
  • WOOF! has a strong track record of delivering value to Compound (new assets, TVL growth, forum engagement, governance participation), making them a credible team to execute this.— Pasha
  • Novel mechanisms like the PID controller, Sandbox Community Pool, and position migration create a flywheel that benefits creators, users, and the DAO simultaneously.— Pasha, dmitriywoofsoftware
  • Sandbox offers a lower-friction path for new projects to prove viability and eventually graduate to the main Compound UI, reducing governance overhead for market creation compared to the traditional proposal-vote-execution cycle.— Pasha, Tane, dmitriywoofsoftware
  • The proposed cost and 24-week timeline seem reasonable and timely given the scope of the project.— PGov
  • The proposal is well-thought-out and could meaningfully boost Compound's TVL and exposure, especially with added transparency features like the Risk Score.— AranaDigital

Arguments against

  • Permissionless money markets have historically ended in disaster (e.g., SushiSwap Kashi, Rari Fuse, Ajna), raising doubts that Sandbox will succeed where others failed.— cmrn, misher
  • Using 'permissionless' and 'whitelist' together is contradictory, since oracle and UI whitelists mean the system isn't truly permissionless.— cmrn
  • The $600K cost, and especially the $300K paid upfront to WOOF! before development begins, is excessive and was not adequately scrutinized by the community.— mexilc
  • The proposal's on-chain execution only funds development without enforcing timelines, deliverable quality, or security requirements, and the upfront $300K transfer is not recoverable if development stalls.— jbass-oz
  • Introducing a lower-quality, higher-risk product alongside Compound's main markets could damage Compound's reputation for safety and reliability.— misher
  • Rather than building permissionless infrastructure, Compound should simply accelerate listing of promising assets that already meet criteria, since the broader DeFi market already provides space for experimentation.— cmrn

Unresolved concerns

  • A detailed breakdown of the $600,000 budget was requested but never provided, leaving its reasonableness unverified.— Tane, mexilc
  • No clear contingency plan exists if Sandbox fails to reach the projected $1 billion TVL target.— jengajojo
  • Suggestions that market creators undergo KYC and that risk disclaimers be shown to users were raised but not addressed by the proposers.— misher
  • Implementation details of the PID controller and how 'experimental' markets can be trusted without curator management remain unclear.— Tane
  • Concerns about oracle reliability for assets not covered by Chainlink/RedStone (e.g., reliance on Uniswap TWAP) were raised without a clear resolution.— cmrn

Notable participants

  • PashaProposer from WOOF! who authored the proposal, responded to community questions, and posted on-chain voting updates.
  • dmitriywoofsoftwareWOOF! team member who provided an FAQ and detailed technical/design responses to community concerns.
  • cmrnMost vocal critic, repeatedly challenging the permissionless model, whitelist contradictions, and value proposition.
  • misherSkeptical commenter who shared a personal loss from a similar permissionless protocol and raised reputational and KYC concerns.
  • mexilcRaised concerns about the high cost and large upfront payment to WOOF!.
  • jbass-ozRepresented OpenZeppelin's review, flagging execution risks around the upfront $300K transfer and lack of enforcement mechanisms.
  • TaneProvided balanced feedback supporting governance efficiency benefits while requesting budget transparency and PID controller details.
  • jengajojoAsked about competing market interests and TVL contingency plans.
  • PGovExpressed support and compared Sandbox to Morpho's model.
  • AranaDigitalSupported the proposal while recommending additional risk transparency measures.
Generated by Kvorum
model: claude-sonnet-5
provider: anthropic
profile: anthropic-strong-v1
prompt: v1.0
generated: 2mo ago
Open the thread ↗