[SEP #7] Governance Framework

gm great start! glad to join

I took a first pass at the draft leaving questions, comments, and suggestions in the Google doc. I included some of the main points here as doc comments should be expected to be resolved and hidden.

Nice work on this so far by everyone! :pray:t2:

Safe Ecosystem Foundation (SEF) and SafeDAO relationship


This is awesome as it would be a big step forward for DAOs by showing a successful legal structure for other serious DAOs to take moving forward.

DAO member liabilities public clarification

It would be beneficial to define the current state (before committees are established) in terms of how participating DAO members who are not contractors or full-time employees of SEF are or are not covered under limited liability with Swiss Foundations. Then, compare this with how DAO members interact with SEF as a result of creating the sub-committee structure.

From what I understand, under Swiss Associations participating members have limited liability under the association. *Not legal advice.

This paragraph states that the goal is to provide limited liability. Does this mean Swiss Foundations work similar to Swiss Associations in providing DAO members limited liabilities?

SEF voting on SafeDAO proposals

Is this saying SafeDAO and the Foundation have not voted on past proposals and will aim to abstain from voting moving forward?

Voting process

Voting strategies

Ranked choice voting could be great to include. Voters ranking the options may lead to people thinking more in-depth evaluating the benefits and tradeoffs of each option rather than focusing mainly on one solution.

As @jengajojo mentioned above, it would be interesting to explore if and how Quadratic voting has helped in the governance process of other DAOs.

Encourage more public communication from delegates on voting decisions

I’ve reached out to a delegate in the past to understand their thought process before. It would be great to have this be public when possible.

Balance of debate between delegators and delegates

Do you think there is room for healthy (constructive and well intentioned) debate between delegates and delgators?


It’s important to clearly communicate what seasons mean externally.


governance system has lots of potential. Yet somne has peter pan syndrome and goes backwards rather than forward.

Kudos to @Andre & @Christoph (and everyone behind the scene) for the dedication and thoughts they’ve put into creating this governance proposal. I’m really excited to see how our governance framework is taking shape - The timelines outlined for each season, combined with the commitment to continuous improvement are awesome!

I love that there’s the review period specifically for the framework amendment - it’s as if we’re engineering a self-iterating system. And yet, as we all know, there’s always room for a bit of fine-tuning.

I’ve got some questions and thoughts I’d like to share. I believe these ideas could make our processes even clearer, more align with the goal, and, most importantly, more inclusive! Because let’s remember, we’re not just crafting something cool here - we’re piecing together a representation of us all.


I feel that it would be great if you could articulate the difference between Delegates and Safe Guardians. My interpretation is that Safe Guardians are akin to what other DAOs term as “recognized delegates”, carrying significant weight as they possess the power to shift proposals into the voting phase. Whereas anyone could be delegates (but not recognized).

That said, I really look forward Guardians V2, also I think it would be helpful to ensure there are checks and balances in place to account for the actions (or inaction) of these guardians. Having like a “accountability checkpoint” for our Guardians and to update their status (whether they’re active or not) would be crucial and i think we incorporate this into our consideration for V2

Governance Cycles

On the topic of governance cycles, I believe we might benefit from a little more agility during our initial phases. Given that we’re embarking on a journey of “governance experimentation”, a 4-5 months cycle might feel a tad long, particularly for our first couple of seasons.

As an alternative, we could adopt a three-sprint cycle approach, with two sprints dedicated to the governance scope and one for review and amendments. This setup would equate to ~3 months or roughly a over a quarter. This could provide us with a bit more agility, allowing us to pivot and adapt as things evolve. Let me know what you think!

Telegram group

I have some thoughts on the Safe Guardian Telegram. I think our goal should be on how we could boost transparency and inclusivity. At the moment, it seems that conversations within this channel are largely restricted to Safe Guardians. To foster a greater sense of openness and collaboration, might we consider making these dialogues accessible to a broader audience? I don’t feel it’s right to have this exclusive chat for only “guaridans”.

Proposal Process

While going through this part, I’ve been curious about the reasoning behind the fixed 60,000 SAFE token requirement for signaling approval. Wouldn’t it be more adaptable and future-proof to link this requirement to a percentage of the SAFE votable supply, rather than a fixed amount? With a percentage-based threshold, our process could better reflect the evolving stakeholders and remain flexible over time. It would also promote the guardians and delegates even more to work towards contributing to safe ecosystem for voting power if they want to keep this privilege.

Additionally, it would be valuable to know how many Guardians and Delegates currently have voting power over the proposed threshold. This information would help us gauge the potential influence spread within our governance process.

Voting Power

This raises an interesting question - what is the current votable supply for SafeDAO? I know we had some data on voting power dynamics, but that’s almost a year old now.

Given all the changes and developments, especially the grants program and the upcoming wave of SGPs, I suspect the landscape has shifted quite a bit. I’m planning to dive deeper into these numbers to create an updated picture, and I’d appreciate any insights from others who have been keeping track.

Minor clarification

I’d like clarification. Does this include only holders, or does it also extend to Delegates and Guardians with voting power above this threshold?

Looking forward to hearing your thoughts and continuing our collective effort to refine and improve our governance framework.


This is a great proposal and love to see the amazing progress that was made here.

There are two trends I see in Governance today which I think Safe could experiment and incorporate in the future. It might still be early but let me drop these in here for future reference.

  1. Stakeholders recognition.
    There are several constituents in the Safe DAO. Currently all are lumped in together, and incentivized under the same ‘Token’. In the future, each such stakeholder will have different perspective of the future, needs, incentives, and potentially, even voting power allocations.

    • Team & Guardians (Those who work for Safe DAO)
    • Safe Users (Those who pay gas, and perhaps fees, to the Safe Smart contracts)
    • Safe financial contributors (Investors, and general token holders looking to maximize the value of the token)
  2. Delegation –
    The general trend in Governance is delegation and would love to see how Safe applies and scales on this primitive. Safe Guardians are an amazing first step, and would love to also see how delegates become even more formalized with responsibilities, accountabilities, and incentives with the general notion that Delegates are a foundational to the DAO.

Once back from ETHcc, I will share a more formalized version of these ideas.


Hi everyone, I’ve been following this conversation and I’m impressed by the depth and breadth of thought that’s been invested into this proposal. I have a few thoughts I shared in the Governance Discord call that I’m posting here for good measure:

On the topic of delegates being actively involved in governance, I wholeheartedly agree. I believe that further steps could be taken to incentivize and ensure active participation of delegates and generate more interest in SafeDAO governance. See 1inch’s Recognized Delegate Program and Aave DAO’s Incentivized Delegate Campaign for examples of delegate compensation models and its effects. This could actually work in conjunction with The Guardians program, wherein a next version could potentially be comprised of the most active existing delegates, and would be a great addition to a SafeDAO Resource Allocation Model. A further step could be, as a community member mentioned in the thread, implementing measures to mitigate the risk of sticky delegation, such as regular delegate re-election or re-delegation campaigns. See Uniswap Foundation’s Delegate Race and Optimism’s Delegate Discovery Initiatives for inspiration.

Regarding the statement that “SafeDAO is stewarded by the Safe Ecosystem Foundation, consistent with the SafeDAO Constitution and the Safe Ecosystem Foundation Charter”, and that “the Foundation acts as a legal wrapper of the SafeDAO treasury, facilitates and administers the governance procedures and implements passed governance proposals if necessary”, I understand the need for a legal entity like the Safe Ecosystem Foundation. However, I agree with other community members who have suggested that the wording here might unintentionally communicate a top-down governance model. Perhaps we could revise the phrasing to better relay the stewardship role of the Foundation, and the agency of SafeDAO and its members.

The concept of “Seasons”, where SafeDAO will change its decision-making process over a specified time frame and run several governance experiments, is pretty interesting, and I see its potential for allowing ongoing iterations and improvements in the governance process. However, further explanation on how a Season is defined and what it encompasses would be beneficial for clarity. A dedicated Glossary or Definitions section in the proposal could serve to delineate such concepts. Maybe we can actually add “Definition of Terms” as a section in the proposal template, encouraging proposal authors to provide additional context and meanings for terms they use.

Lastly, on the points about delegate and token holder participation, this highlights an important aspect of delegate-token holder relationship dynamics. To further clarify and formalize this relationship, I propose we design a “Delegation Agreement”. Delegates could adopt this agreement to define their commitment and decision-making processes in permissionless delegation or state their preferred terms. This, coupled with clear delegate statements, can help provide more transparency in this dynamic.

Some other things I think could be nice to include in a final version of this prop:

  • An org chart depicting stakeholder groups within SafeDAO & showing how their responsibilities will overlap or diverge
  • Clarify what happens if there isn’t sufficient participation in a voting sprint
  • Share how the specifics of voting systems for proposals are decided or could be. Case study examples maybe?

Overall great work, and look forward to seeing the next version of this!


Great points @v3naru, @Eylon, and @Nneoma_StableLab!

Difference between Delegates and Safe Guardians

This is my understanding.

  • Guardians: Stewards of SafeDAO in their given area of expertise, e.g. Developers, designers, product, governance, and etc. They provide their energy to helping the DAO, core Safe team, developers, etc. in their given domain.
  • Delegates: Specifically focused on creating and analyzing existing governance proposals, communicating their perspective to the ecosystem and guiding proposals in the direction they see best for the ecosystem.

Guardian checks and balances

This is an interesting idea. It could be useful to have a dynamic onboarding/offboarding system.

  • Self-regulating
    • For example, say Q1 2024 I’m planning to spend the majority of my energy on another project, lower than the determined threshold for the SafeDAO. I may indicate to SafeDAO I will not be an active Guardian during this time period. If there are associated rewards, they would not apply to me during Q1 2024.
    • Then, Q2 2024, I plan to rejoin as a Guardian and allocate the required threshold of time to SafeDAO. I indicate my intent to SafeDAO, and I am potentially approved through voting to rejoin.
  • Community
    • The community could check-in with Guardians that are inactive to see if they are planning to remain active for the upcoming quarter.
    • Hopefully, Guardians would be self-aware enough to self-regulate to inactive if they cannot spend the energy required. However, there could be a community mechanism in more extreme cases where Guardians become inactive without self-regulating.

Broader audience for communication

Moving towards realtime and casual discussion on public places like Discord (#governance channel) or other more opensource platforms seems like a good move compared to closed Telegram channel.

Dynamic SAFE token requirement to signal approval

This makes sense as token distribution will change post-transferability.

Additions to framework

  • Definition of terms: This will help improve clarity and onboard new ecosystem members.
  • Delegation agreement: Defining clear expectations is important.

Hello, can you please tell me where I can read the requirements and specifics Hopefully?

Thanks everyone for the comprehensive feedback so far!

A lot of great comments and ideas are brought up in this thread, the Google Doc, and in both governance calls. We are currently in the process of triaging the feedback to determine its incorporation into the initial SEP and identify ideas for future governance iterations.


Just wanted to drop a quick note to let you all know that we’re still actively incorporating the recent feedback we’ve received. We plan to share a revised version of the SEP later this week.

Additionally, please remember that we have a community call scheduled for next week. All the details can be found here: Community Call #11.

1 Like

[Moving proposal from Phase 0 to Phase 1]

Thank you everyone who provided their insights and feedback to the governance framework! Across the forum post, Google doc and the governance calls we received over 100 comments and identified over 50 distinct feedback topics.

We’ve thoroughly reviewed each one and integrated much of your feedback. We have made an effort to acknowledge everyone who contributed (please see the section at the end “Acknowledgements”), but if you feel you’ve been overlooked or that your feedback hasn’t been adequately addressed, please let us know. We aim to ensure that no contribution is missed and that all voices are recognized.

  • A consolidated list of all feedback, along with responses if it was incorporated or not at this stage, can be found here.
  • The most recent version of the governance framework can be found in the second post of this thread and here as a Google Doc.
  • Additionally, you can compare the iterations between Phase 0 and the current Phase 1 version here.

With these updates, we’re excited to advance this proposal to Phase 1.

Many of the comments we’ve received touched on future governance experiments (e.g. additional voting types) or resource allocation (e.g. incentivizing delegates or guardians v2). As such they are not part of the initial governance framework.

We’ve also received several inquiries about the Foundation’s function and its role in limiting liability for SafeDAO. To provide context:

Origin and function:

The Foundation was implemented as part of Gnosis spin-off proposal (GIP-29: Spin-off safeDAO and Launch SAFE Token - GIPs - Gnosis). It acts as the steward of SafeDAO. The Foundation’s goal is to provide an innovative Swiss solution to the legal challenges of DAOs, such as lack of legal personality, the limitation of DAO participants’ liability, and the tax treatment of community assets by leveraging SEF’s legal personality in engagements with contracting parties and acting as a legal wrapper of the treasury for tax purposes.


Checks and balances are in place as the Foundation is

  • legally bound by its Deed,
  • the purpose of the Foundation (the full text of the purpose of the Foundation in its legally binding, German version can be found here. An unofficial translated version (German / English) can be found here.) and
  • is under supervision of ESA to foster the Safe ecosystem.

Relation to SafeDAO

The Foundation and SafeDAO are two separate entities. To increase the interaction and control, the Foundation and SafeDAO can utilize the native bodies of DAO committees to interact closer and delegate responsibilities to DAO members constituting these DAO committees. Currently, SEF is clarifying the organizational requirements for the establishment of DAO committees with the relevant authorities. Once these requirements have been clarified, it is intended that SEF and SafeDAO will jointly coordinate the establishment of DAO committees.

KYC requirements

Some queries also revolved around KYC. KYC is required to be part of DAO committees and to receive resources from SafeDAOs treasury administered by the Foundation. We are currently evaluating the details of this and which service providers can be used to streamline this process.

As we gain more legal clarity on the DAO committees, we are looking to add more info to the Foundation section of our SafeDAO Governance Hub and will also release a dedicated blog post.


Hey @Andre ,should we move to the next stage given that this thread hasn’t been updated in two weeks

1 Like

Before we proceed to the voting phase, I would like to give everyone the chance to check the [SEP #X] Outcomes-based resource allocation framework (OBRA) we recently posted as a draft in Phase 0, as it directly relates to the governance cycles of our governance framework (refer to sections D.II “Governance cycles” here and B. “Submission and review cycle” of the resource allocation framework).

For visibility I’m also adding here that today we are hosting a governance call on the resource allocation framework:
:calendar: Tuesday, Sept 5 at 7pm CET in the ⁠governance-discussions channel on Discord
:point_down: RSVP/Add to cal: Discord


Hi all, we’ve refined the governance framework to incorporate insights we identified over time and from discussions surrounding the resource allocation framework.

In addition to minor linguistic modifications, the following key changes have been made:

  • The duration of sprints has been corrected from 5 to 4 weeks. This adjustment was made after reviewing the timeline and also aligns with the feedback in favour of more iterations within a year.
  • We have now clearly stipulated an exception, allowing voting on the milestones of SEP #3 to occur in any sprint. Given the predictable nature of these milestones, the need for streamlining them through designated governance and non-governance sprints is not needed. While this was previously also possible due to the soft launch, the amendment serves to make this clear.
  • As a result of this change, this allows the governance framework to enter into force shortly after the ratification (with a small delay to prepare the kick-off), moving away from the previous dependency on transferability of the Safe token.

You can find all changes reflected in the first two initial posts as edits and also documented with track changes in this Google Doc:

[A remark on the interrelation to the resource allocation framework:
The governance framework does not address or anticipate any connection with the resource allocation framework yet, as this is a distinct SEP, which needs to be voted on seperately. Given that OBRA has governance-related components, it will amend some parts of the governance framework, e.g. add new proposal types for amending strategies or terminating initiatives. As such, the resource allocation framework will serve as an amending act to the governance framework. The detail of these changes will be posted in the respective thread for OBRA as part of the next draft.]

Through these edits, the review period has been extended in accordance with the current governance process. We aim to finalize the governance framework and proceed to a vote after the review period of a minimum of 6 days has ended - provided there are no critical comments on these amendments.


We’ve made a minor revision to the wording under V.2. Voting formalities:

Who: The Safe Ecosystem Foundation can add proposals to Snapshot that are eligible to Phase 2 or entrust the responsibility to (a group of) guardians or other SafeDAO participants. In addition, any token holder with at least 20,000 Safe token can technically post proposals on Snapshot.

Under Annex 2, we just added brackets around a group of

The changes are reflected in the initial posts and also documented with track changes in the Google Doc above.

As this change is minor and does not restrict any rights but just opens up the possibility for guardians or other SafeDAO members to take over certain responsibilities, I don’t think this should extend again the review period. However, I will leave this 1-2 days on the forum before proceeding to a vote.


Quick question: Why is stablelab putting this proposal on snapshot?

I asked them if they could upload it for me as an author. If you look at the previous snapshot proposals, authorship and uploading is often not done by the same people. Technically anyone with more than 20k Safe voting power can upload a proposal.


:mega: [SEP #7] Governance Framework has been uploaded to Snapshot. Voting begins today, Friday, at around 5pm CEST and the voting period is open for 7 days.

I would ask everyone to prepare for voting or reach out to your delegate. If you run into any issues or have questions let me know.


:ballot_box: Vote on Governance Framework [SEP #7] has started! The voting period is open for 7 days.

Cast your vote or reach out to your delegate: https://snapshot.org/safe.eth#/safe.eth/proposal/0x71bf4469fc289bab2d8ce99ba41d5792c1c5d581defc12042d3cf1eb3e136323


@Andre is this about token transferability?