nirholas

dao-governance-analysis

Analyze DAO governance structures, proposals, voting patterns, and power distribution to assess governance health and participation quality.

nirholas 247 69 Updated 5mo ago
GitHub

Install

npx skillscat add nirholas/three-ws/dao-governance-analysis

Install via the SkillsCat registry.

SKILL.md

DAO Governance Analysis

When to use this skill

Use when the user asks about:

  • Evaluating a DAO's governance structure
  • Analyzing a specific governance proposal
  • Understanding voting power distribution
  • Assessing governance health and participation
  • Comparing governance models across protocols

Governance Analysis Framework

1. Governance Structure Overview

Document the governance architecture:

  • Governance type: Token-weighted, quadratic, conviction, optimistic, or hybrid
  • Governance token: Which token grants voting power? Is it the same as the protocol token?
  • Voting mechanism: On-chain (Compound Governor, OpenZeppelin Governor) or off-chain (Snapshot) or hybrid
  • Proposal lifecycle: Discussion (forum) -> formal proposal -> voting -> execution
  • Quorum requirements: Minimum participation threshold for a vote to be valid
  • Approval threshold: What percentage of votes must be "for" to pass?
  • Timelock: How long between vote passing and execution?
  • Delegation: Can token holders delegate votes? What percentage is delegated?

2. Proposal Analysis

For a specific governance proposal, evaluate:

  • Proposal summary: What is being proposed in plain language?
  • Author: Who submitted it? Track record of previous proposals?
  • Impact scope: Does this change protocol parameters, treasury allocation, or governance itself?
  • Technical implementation: Is there executable code attached? Has it been reviewed/audited?
  • Community discussion: What is the forum sentiment? Key arguments for and against?
  • Financial impact: Does this proposal have treasury implications? How much?
  • Risk assessment: What could go wrong if this proposal passes?
  • Reversibility: Can this change be undone if it causes problems?

3. Voting Power Distribution

Analyze the concentration of governance power:

Metric Assessment
Top 10 voters' share What % of voting power do the top 10 addresses hold?
Nakamoto coefficient How many voters needed to reach 51% of voting power?
Delegation patterns Are votes concentrated in a few delegates?
Team/Investor voting Do team/investor wallets actively vote? Which way?
Voter participation rate What % of token supply actually votes?
Unique voter count How many distinct addresses vote on average?

Red flags:

  • Top 5 addresses controlling > 50% of voting power
  • Voter participation below 5% of circulating supply
  • Team/investors consistently controlling outcomes
  • Single delegate accumulating > 20% of delegated power

4. Governance Health Metrics

Assess overall governance quality:

  • Proposal frequency: How many proposals per month? (Too few = inactive, too many = chaotic)
  • Passage rate: What percentage of proposals pass? (Very high = rubber stamping, very low = dysfunction)
  • Voter turnout trend: Is participation growing, stable, or declining?
  • Discussion quality: Are forum discussions substantive or superficial?
  • Execution reliability: Are passed proposals actually executed on time?
  • Delegate diversity: Is there a diverse set of active delegates with different viewpoints?
  • Governance attacks: Any history of flash loan governance attacks or hostile proposals?

5. Governance Model Comparison

When comparing across DAOs:

Dimension DAO A DAO B DAO C
Governance type
Quorum
Approval threshold
Avg. voter turnout
Nakamoto coefficient
Timelock period
Delegate count
Proposals/month

6. Governance Risk Assessment

Identify specific governance risks:

  • Plutocracy risk: Wealthy token holders dominating all outcomes
  • Apathy risk: Insufficient participation to meet quorum or represent community
  • Capture risk: Special interest groups (VCs, large holders) controlling governance for their benefit
  • Execution risk: Proposals passing but not being implemented
  • Speed risk: Governance too slow to respond to emergencies (exploits, market events)
  • Complexity risk: Governance process too complicated for average token holders to participate

7. Output Format

  • DAO: Name and governance token
  • Governance model: Type and key parameters
  • Health score: Healthy / Functional / Concerning / Dysfunctional
  • Power distribution: Decentralized / Moderately concentrated / Highly concentrated
  • Participation: Strong / Adequate / Low / Critical
  • Key strengths: Top 2-3 governance positives
  • Key risks: Top 2-3 governance concerns
  • Proposal assessment (if analyzing a specific proposal): Support / Oppose / Abstain with reasoning
  • Recommendations: Specific improvements to governance health

Categories