HSP-XXX: Creation of a Due Diligence Team (DDT) under the House of Stake Security Council

hsp: 004

title: Creation of a Due Diligence Team (DDT) under the House of Stake Security Council
description: Establish an independent due diligence unit to enhance integrity, transparency, and conflict-of-interest auditing across HoS governance operations
author: @Dacha
discussions-to: HSP-XXX: Creation of a Due Diligence Team (DDT) under the House of Stake Security Council
status: Proposal
type: Sensing
category: Governance Oversight
created: 2025-10-12

Abstract

This proposal creates an independent Due Diligence Team (DDT) operating under the House of Stake (HoS) Security Council.
The DDT will conduct conflict-of-interest (COI) and integrity reviews of proposals, budgets, appointments, and partnerships across HoS operations.
Its mandate is to ensure independent review, transparency, and risk mitigation, and to publish actionable recommendations for the Security Council.

Context & Alignment

In recent governance cycles, multiple HoS discussions raised concerns about potential conflicts of interest, unclear decision processes, and limited risk auditing for grants, appointments, and partnerships.
Currently, no formal independent body performs integrity checks before proposals advance to vote.
The DDT introduces an audit layer aligned with the HoS Constitution principles of transparency, accountability, and fairness, and mirrors best practices used in other ecosystems (Arbitrum, Optimism, Cosmos).

This proposal does not modify NEAR protocol rules or smart contracts; it establishes an organizational functionwithin HoS governance to improve procedural integrity.

Situation

House of Stake has grown rapidly, processing multiple funding, partnership, and structural proposals.
Without a standardized audit mechanism, potential self-dealing, double-funding, or undisclosed relationships may undermine trust and decentralization.
The Security Council currently lacks a dedicated team to perform background checks or integrity analysis.

The DDT addresses this gap by providing an independent pre-vote assessment of major proposals and actors within the governance process.

Mission

Objectives

• Establish a three-member independent DDT under the Security Council

• Conduct COI reviews and background checks for all proposals.

• Deliver written reports to the Security Council within 7 days of assignment

• Maintain a public log (monthly summary) of audits completed

Measurable Outcomes (Q1 2026 Pilot)

Approach

Program Structure (Pilot Phase: Q1 2026)

Mandate:
The DDT independently reviews and reports on integrity, conflict-of-interest, and governance risks.

Composition:

• 3 members appointed by the Security Council

• Experience in NEAR governance + at least one external L1/L2 ecosystem

• No paid or contractual conflicts (NF, HoS Ops, related to the NF contractors)

Compensation:

• Roles are primarily volunteer; task-based one-off payments may be authorized by the Security Council

• Budget requests submitted via Screening Committee → HoS vote (without alteration)

Reporting Line:
DDT → Security Council → Publication

Operating Procedure

1. Intake: Receive all proposals flagged for Council review.

2. Initial Screening: Assign reviewers by area of expertise.

3. Fact Gathering: Request clarifications or background data from proposers.

4. Report Drafting: Include clear recommendation (Approve / Approve w Changes / Defer / Reject).

5. Submission: Deliver final report to the Security Council.

6. Publication: Council decision and (where possible) public summary released.

Future Expansion (Beyond Pilot)

If effective, the DDT may evolve into a permanent HoS Integrity Office with rotating membership, subject to new proposal approval.

Technical Specification

• Data Sources: on-chain wallet analysis, prior HoS proposals, NF grant databases, public records, and KYC attestations (if required).

• Reporting Format: standardized PDF/Markdown form including COI mapping, related wallets, prior grants, and risk rating (Low / Medium / High).

• Storage: reports archived in HoS GitHub and linked in governance portal.

• Access Control: raw data private to Security Council; public summaries for transparency.

Backwards Compatibility

No impact on NEAR protocol or existing smart contracts.
Adds a new off-chain process within the HoS governance framework.
Existing governance proposals remain valid; only the review pipeline changes.

Implementation Timeline

Budget & Resources (Pilot Quarter)

Team & Accountability

Responsible: House of Stake Security Council
Accountable to: House of Stake Governance Community

Roles:

• Security Council – appoints, supervises, approves budget

• DDT Members – conduct audits and publish reports

• Community – reviews public summaries and feedback

Security Considerations

• Governance Integrity: reduces risk of self-dealing, double-funding, or opaque appointments.

• Independence: eligibility and compensation separation prevent capture by operational entities.

• Data Security: sensitive findings handled under restricted access; sanitized summaries shared publicly.

• Abuse Prevention: public logs deter misuse; Council oversight ensures accountability.

Conflict of Interest

• I have read and agree with the House of Stake Conflict of Interest Policy

Copyright

Copyright and related rights waived via CC0 1.0 Universal.

• I think the community should be the one identifying conflicts of interests and bringing them up in discussions or on-chain platforms (not sure if there’s a process for it yet). If there is a victim of a conflict of interest, it will be in their interest to report it
• “Pre-Vote Reviews: Advise the Security Council on whether proposals are safe” - The security council works post-vote. Pre-vote is pre-screening committee.
• I like to keep the number of gears as small as possible to avoid friction, additional delays, and division of responsibilities (which often results in neither side taking their job seriously because they think the other side will take / has taken care of it). Currently, due diligence is the responsibility of security council, and the voters who need to make their decision based on public facts.

slimedrgn:

I like to keep the number of gears as small as possible to avoid friction,

Agree. NDC had so many moving parts and nobody was clear what decisions must be made by whom.

I would say HoS is still too many moving parts rn. We should aim to simplify - all onchain, directly on top of voting proposals (+near.social for comments). Pre-screening done by AI agent that can evaluate mandate alignment.

Security council minimally involved to track technical issues with contracts and upgrades.

And to minimize future conflict of interest - let’s get AI delegates up and running where everyone can inspect what is their prompt/decision making.

illia:

let’s get AI delegates up

Yes, that’s the North Star. I’d really like to see such agents live by the end of the year. Unfortunately, no one is publishing progress updates, so it’s unclear where we currently stand.

Regarding the commission — someone must handle conflict-of-interest checks. There are volunteers for this, that’s great; otherwise, we’ll repeat the mistakes of the NDC and the reasons it was so heavily criticized.

This body does not complicate HoS processes in any way. It is purely informational for the Security Council and costs nothing.

hsp: 003

title: Due Diligence Team under Security Council
description: Establishes an independent Due Diligence Team under the Security Council to audit conflicts of interest, increase transparency, and provide governance recommendations.
author: Dacha
status: Proposal
type: Governance / Operational
category: Governance / Oversight
created: 2025-10-12

Abstract

This proposal establishes an independent Due Diligence Team (DDT) under the House of Stake (HoS) Security Council to strengthen integrity, transparency, and risk management within HoS governance.

The DDT will:

• Review potential conflicts of interest (COI)
• Verify backgrounds and track records of proposers, vendors, and appointees
• Produce due diligence reports with clear recommendations
• Advise the Security Council before proposals, budgets, or appointments move to a vote

By introducing an independent audit layer, this proposal increases trust, reduces self-dealing risks, and aligns HoS governance with best practices from leading Web3 ecosystems.

Context & Alignment

The House of Stake Constitution and NEAR values emphasize:

• Transparency
• Accountability
• Decentralization
• Community trust
• Responsible treasury stewardship

However, current governance flows rely heavily on internal committees (e.g., Screening Committee), which may face conflicts of interest or lack independent oversight.

Other L1/L2 ecosystems (e.g., Arbitrum, Optimism, Polygon, Cosmos) have introduced checks and balances such as risk committees, COI policies, and external audits to maintain governance legitimacy.

This proposal aligns HoS with these industry best practices by creating an independent Due Diligence Team under the Security Council.

Situation

As HoS governance expands—more budgets, proposals, appointments, and vendors—the risk of conflicts of interest increases. Examples include:

• Self-dealing or favoritism in budget allocation
• Undisclosed relationships between proposers and service providers
• Lack of background verification for appointed roles
• Screening Committee making decisions without independent oversight

Current issues:

• No dedicated body to detect or evaluate conflicts of interest
• No formal due diligence process
• No independent review before proposals reach governance vote
• Community concerns about transparency and accountability

Without a COI audit layer, HoS risks repeating failures seen in other ecosystems (e.g., NDC, DAO controversies) and losing community trust.

The DDT resolves this by providing structured, independent review and recommendations to the Security Council before decisions are finalized.

Mission

Objectives

The Due Diligence Team (DDT) will:

• Strengthen integrity, transparency, and trust within HoS governance
• Identify and flag conflicts of interest (COI) in proposals, budgets, partnerships, and appointments
• Verify the background and track record of proposers, advisors, vendors, and sub-DAOs
• Ensure proposals are aligned with NEAR values and HoS Constitution
• Improve governance decision quality through clear, unbiased recommendations
• Provide independent oversight over internal governance bodies (e.g., Screening Committee)

Outcomes

• Reduced self-dealing and hidden conflicts of interest
• Increased transparency in decision-making and budget flows
• Stronger accountability for all actors within HoS
• Higher community trust and greater legitimacy of governance outcomes
• Alignment with industry best practices for DAO governance and risk management

Team Composition

The composition of the DDT is determined and appointed exclusively by the House of Stake Security Council.

The DDT consists of 3 members, each with a combination of:

1. Proven practical experience inside the NEAR ecosystem, such as:

   • Governance
   • Treasury or budget management
   • Community or operations leadership
   • Proposal evaluation or compliance

2. Hands-on governance track record in at least one other major Web3 L1 or L2, e.g.:

   • Arbitrum
   • Optimism
   • Polygon
   • Cosmos
   • Other established DAO ecosystems

This combination ensures deep understanding of NEAR-specific context and broader industry best practices.

Eligibility Requirements

To ensure independence and avoid conflicts of interest, DDT members must NOT hold any paid operational role that creates a direct conflict within:

• House of Stake Core / Operations
• NEAR Foundation / HoS Foundation
• HoS service providers or contractors
• Any entity that may be subject to DDT review

This guarantees the DDT can investigate any actor, including the Screening Committee or contractors, without bias.

Core Responsibilities

1. COI Audits

• Screen proposals, budget requests, partnerships, and appointments for:
  • Hidden financial ties
  • Double spending
  • Self-dealing or favoritism
  • Undisclosed relationships

2. Background Verification

• Check track record and reputation of:
  • Proposers
  • Advisors
  • Vendors / contractors
  • Sub-DAOs or affiliated entities

3. Pre-Vote Reviews

• Advise the Security Council before proposals advance to governance vote
• Recommend revisions or mitigations when necessary

4. Due Diligence Reports

Each report must include:

• Summary of findings
• Identified risks
• COI assessment
• Recommendation:
  • Approve as-is
  • Approve with changes
  • Defer (needs more information)
  • Reject (with explanation)

5. Governance Process Improvement

• Recommend updates to governance flows to reduce future COI risks
• Provide suggestions for transparency, accountability, and fairness

6. Periodic Transparency

• Maintain a log of reviews performed and concerns flagged
• Share full or sanitized reports with the community when appropriate

Operating Procedures

1. Proposal Intake

• DDT receives all proposals designated for Security Council review

2. Initial Screening

• Members divide fact-checking and COI analysis based on expertise
• Identify obvious red flags or areas needing deeper investigation

3. Fact Gathering

• DDT may:
  • Request clarifications from proposers
  • Request disclosures of relationships or incentives
  • Gather ecosystem references or background information

4. Draft Report

Each draft includes:

• Summary of findings
• Identified conflicts
• Risk level
• Recommended action (approve / approve with changes / defer / reject)

5. Internal Review

• DDT members review each other’s work
• Achieve clear, unbiased final recommendation

6. Submission to Security Council

• DDT submits final report to Security Council for consideration

7. Council Decision

The Security Council may:

• Accept the recommendation
• Amend the recommendation
• Reject the recommendation

8. Publication

• Final decision and outcome are published with minimal delay
• Full or sanitized report may be shared to ensure transparency

Term & Replacement

Term Length

• DDT membership is not strictly time-limited but is renewable with Security Council approval.

Removal

Members may be removed by Security Council vote in case of:

• Conflict of interest
• Breach of trust or confidentiality
• Non-performance or ethical violations

Vacancies

• Filled through the same nomination and vetting process by the Security Council.

Compensation & Budget Structure

Principle

These are not full-time salaried roles.

Discretionary Compensation

If a task requires deep investigation, the Security Council may approve one-off compensation per deliverable.

Budget Flow (to ensure independence)

1. Security Council pre-approves budget
2. Screening Committee must submit proposal as-is
3. HoS governance votes on final approval

This prevents Screening Committee interference and preserves DDT independence.

Reporting Lines

DDT reports directly to the Security Council.

The Security Council may:

• Accept, amend, or reject DDT recommendations.

Final authority remains with the Council.

Governance Integration

Before a proposal goes to vote:

1. Screening Committee drafts proposal
2. Security Council forwards to DDT
3. DDT conducts review
4. DDT submits report
5. Council decides whether proposal advances

No proposal with major conflicts reaches vote unnoticed.

Transparency & Accountability

• DDT maintains a log of reviews and findings.
• Reports may be public or sanitized.
• Final decisions are published promptly.

Why This Matters

This proposal introduces the first independent audit layer within House of Stake.

It will:

• Prevent self-dealing and hidden COIs
• Add transparency to budgets and appointments
• Hold Screening Committee and contractors accountable
• Strengthen community trust and governance legitimacy
• Align HoS with best practices from Arbitrum, Optimism, Cosmos, etc.

Without independent due diligence, HoS may repeat past failures (e.g., NDC).
With DDT, HoS becomes one of the most transparent governance systems in Web3.

Milestones

• DDT Formation: Immediately after approval
• Process Setup: Within 1–2 weeks
• Initial Reviews Begin: After setup
• First Transparency Log: End of Month 1
• Quarterly Audit Reports: Every 3 months
• Annual Review: 12 months after launch

Reporting

Quarterly Reports

• Proposals reviewed
• Conflicts identified
• Recommendations made
• Council decisions
• Improvements proposed
• Any compensated investigations

Annual Review

• Deep performance evaluation
• Effectiveness of DDT
• Systemic risks
• Recommendations for continuation or improvement

Security Considerations

Governance Security

Prevents power concentration
Independent oversight

Conflict of Interest Security

No paid operational roles
Mandatory disclosures

Economic Security

Prevents treasury misuse
Minimal cost, high ROI

Abuse Prevention

Council retains final authority
DDT cannot veto
Members can be removed if compromised

Conflict of Interest

I’ve read and agree with the House of Stake Conflict of Interest Policy. This proposal establishes a structure to prevent and mitigate COIs. The DDT itself will comply with COI standards and maintain independence.

Copyright

@Dacha

Disclaimer

This proposal establishes an advisory oversight mechanism to improve transparency and conflict-of-interest prevention within HoS.

It does not modify the NEAR protocol or validator mechanics.

Final decision-making authority remains with the House of Stake Security Council.

illia:

to minimize future conflict of interest - let’s get AI delegates up and running where everyone can inspect what is their prompt/decision making.

Please stop that AI delegates narrative or explain what exactly it’s supposed to look like (if you have a real plan / specific idea), a lot of people who work with HoS (both delegates and contractors) are blindly following your vision and when asked, are unable to explain why, and I think your vision is wrong here. It’s possible that without my feedback 25-50% the 6-month “budget” (a funny doc that delegates and nf contractors have been writing in one call, I literally had to join while I was sick because most people there don’t know what they’re doing and are just going with the AI narrative) would’ve been allocated to development of AI delegates and 0% would’ve been used for DeFi. “AI delegate” is an experimental gimmick that I think will not work in the next 3-4 years, based on my experience of trying to build semi-autonomous agents and not releasing any of them because they suck. The limitations of context, models with very weak datasets (even commercial ones like OpenAI / Claude are easy to predict and manipulate, and hard to make them act on their own, analyzing past mistakes and experience, they’re just programmed to be valuing some things over the others). AI will also definitely have conflict of interest, AI voting against AI development would make no sense, and phrases like “don’t vote for AI just because you’re an AI“ will be hard to balance, it might just never vote for AI things after that. Prompt engineering will take a long time, and with a dataset of maybe 20 proposals made until the end of the year, it will be impossible to make it reliable. Even in Cursor, sometimes I see wild behavior because of some stupid mistake in a prompt. It usually gets fixed in a few days / weeks, but shows how hard it is to program a balanced and well-behaving agent even for a team with 9 figure revenue, where even 1% makes a huge difference. They’re not even claiming it to be fully autonomous, or very dependable, everyone knows and accepts that sometimes it wonders in a whole different direction. In a constantly evolving crypto space, where frontier models still don’t know about Hot Wallet’s existence (without search tools), it’s even less viable, the models don’t know current community priorities, hype topics, red flags, unless you list them all in the prompt and keep it updated every week. For example, Intear AI Moderator bot in almost all cases treats messages like “Thanks to [insert name with link to channel], I made 2x on trading in 1 month, I recommend her investment advice to everyone!!!“ as good messages. With default prompt, or with prompt that includes “Flag all messages that might be a scam“. Unless you add “Flag messages recommending a financial advisor with a link“, GPT-5, GPT-4o, latest Llama models, and any other models will just treat it as a good message, while any entry-level educated human would instantly see the scam, even if put it completely out of context. The tech is just not there yet. And if you limit to open source / open weight models, that’s like going 1-2 years back in time. Another (but smaller) concern is that if agents are open source with an open prompt, anyone can just run their proposal through the model and optimize it for the model until the model agrees with them.

That being said, I would support less ambitious AI tools, like forum + telegram discussion summarizer, AI Delegate-like interface that can roast your proposal / ask you questions to help with clarity of a proposal when writing it, due diligence agent that can find a company’s website, socials, past work, etc.

I’m also not against independent developers building their own autonomous / semi-autonomous AI delegates if they really want to try, as long as they earn the trust of the community / earn their delegated stake, and are not paid by house of stake to do so.

slimedrgn:

explain what exactly it’s supposed to look like (if you have a real plan / specific idea)

Here was a draft of architecture: https://x.com/ilblackdragon/status/1870370654178189561

There are should be different implementations of AI delegate - that is the actually the beauty of congress like system here as it allows for an easy heads-to-heads experimentation.

High level:

• autonomous agent that has a NEAR account
• people (and other agents) can delegate to it veNEAR
• delegators can provide feedback / opinion to it - optionally it weights it base on their veNEAR or based on their social clout or other metrics
• it then votes on proposals (and ideally comments why it voted in different way)

slimedrgn:

“AI delegate” is an experimental gimmick that I think will not work in the next 3-4 years,

We can disagree on timing but I do think it make sense to build now so we can actually test it out. I don’t expect that it will work from the first try - there is clearly now reason why it would. But this would start creating an environment to test the edge cases of such system.

Because you have full transparency of who provided what feedback and how it reasoned / updated it priors and voted - this becomes a great red teaming place with real enough value while having guard rails with real people delegates and delegators to overrule if it’s offtrack.

Probably a more contained “capture the delegate” hackathon make sense initially where people are explicitly trying to break it while devs are patching it. Creating such dataset would be invaluable IMO.

slimedrgn:

The tech is just not there yet. And if you limit to open source / open weight models, that’s like going 1-2 years back in time.

This is where we disagree with timing. Open weight models are prob ~3-4 months behind and the harder problem is tooling and context around them.

slimedrgn:

That being said, I would support less ambitious AI tools, like forum + telegram discussion summarizer, AI Delegate-like interface that can roast your proposal / ask you questions to help with clarity of a proposal when writing it, due diligence agent that can find a company’s website, socials, past work, etc.

My understanding that is the first step on the roadmap

slimedrgn:

I’m also not against independent developers building their own autonomous / semi-autonomous AI delegates if they really want to try, as long as they earn the trust of the community / earn their delegated stake, and are not paid by house of stake to do so.

Sure