giza-protocol-primary-sources-2026-05-11

Giza Protocol primary sources

Collected: 2026-05-11 UTC

Primary sources reviewed

• https://docs.gizaprotocol.ai/
  • Official Giza documentation home
  • Used to confirm the current top-level split between protocol, agents, token, staking, and governance surfaces
• https://docs.gizaprotocol.ai/introduction/protocol
  • Official protocol architecture page
  • Used to confirm Giza’s three-layer framing: semantic abstraction via MCP, ERC-4337-based authorization, and EigenLayer AVS execution
• https://docs.gizaprotocol.ai/introduction/agents
  • Official agent overview
  • Used to confirm Giza’s own positioning of agents as non-custodial autonomous financial entities and ARMA as the first deployed product surface
• https://docs.gizaprotocol.ai/staking/overview
  • Official staking docs
  • Used to confirm the split between operator staking for execution-network security and community staking for governance participation
• https://docs.gizaprotocol.ai/governance/overview
  • Official governance docs
  • Used to confirm Giza’s explicit council-based, delegation-heavy governance roadmap and phased decentralization model
• https://docs.gizaprotocol.ai/token/purpose-utility
  • Official token-utility docs
  • Used to confirm GIZA’s role in operator staking, progressive governance, and liquidity-pair-based fee processing
• https://github.com/gizatechxyz
  • Official GitHub organization
  • Used as repository evidence that Giza maintains first-party protocol, agent, token, and zkML tooling rather than only a hosted app surface
• https://github.com/gizatechxyz/giza-agents
  • Official agents repository
  • Used to confirm the agent framework’s own language around trust-minimized ML-to-smart-contract interoperability and lifecycle/iteration tooling

Notes

• Giza cleared the bar because it is not just another DeFi agent product or generic AI x crypto brand. The strongest analytical split in the docs is its three-layer architecture: MCP-based semantic abstraction, ERC-4337 smart-account authorization, and EigenLayer-AVS execution.
• The semantic-abstraction layer is especially useful for the corpus. Giza is making protocol-specific DeFi interactions legible as a standardized intent surface, which is a different control plane from model inference itself.
• The authorization layer matters just as much as the agent UX. By centering smart accounts, validator modules, and programmable session-key-like permissions, Giza frames agent autonomy as constrained delegation rather than as custody transfer.
• The execution layer makes operator admission and collateral policy explicit. The docs say operators stake GIZA to run the decentralized execution network, which means autonomous-finance promises ultimately depend on who can execute, under what slashing/accountability regime, and how cross-chain batching/messaging are handled.
• Governance is also a first-class comparison surface here. Giza explicitly rejects pure token-holder direct democracy in favor of specialized councils, delegation, and time-weighted influence, which makes the governance layer more comparable to expert-control middleware than to ordinary DAO voting.
• This entry belongs in the active corpus because it gives the crypto × AI cluster a useful comparison point where semantic protocol abstraction, non-custodial delegated authority, execution-operator staking, and phased expert governance are all visible in one stack.