Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.stratareserve.co/llms.txt

Use this file to discover all available pages before exploring further.

Strata’s governance layer bridges human oversight, autonomous agents, and token- holder participation. It is designed for progressive decentralization: start centralized (founding team and agents), earn trust through track record, and gradually delegate authority to on-chain governance.

​
The Risk Committee

The Risk Committee is the central risk-governance body. Its authority scope:
  • Collateral eligibility — which asset classes qualify for srUSD minting; which SPVs are approved
  • LTV and CR parameters — collateral ratios and loan-to-value limits per asset class
  • Issuance limits — system-wide, per-SPV, and per-engine caps
  • Emergency actions — system-wide pause, engine quarantine, oracle overrides
  • Agent policy ceilings — absolute maxima for all agent (Steward) policy parameters
  • Valuation methodology — valuation approaches, haircuts, oracle sources
  • Concentration limits — maximum borrower and asset-class exposure
  • veSTRA token distribution (governed by token holders)
  • Revenue-split changes greater than 5% (requires a governance vote)
  • Entity-structure changes (a board / legal matter)

​
Phased decentralization

1

Phase 1 — Founding team & agents

The founding team and autonomous agents operate the system under explicit policy ceilings, building a track record.
2

Phase 2 — Independent oversight

An third-party risk advisor joins for independent model validation and oversight.
3

Phase 3 — Token-holder representation

A community seat, elected by veSTRA holders, joins governance, expanding on-chain participation.

​
Agents under policy

Autonomous agents execute operations within bounds the Risk Committee sets. Agent policy ceilings define absolute maxima; state-changing on-chain actions are proposed and gated through a policy layer (Steward) before signing. Agents propose; policy and humans sign.
The model is layered: agents act inside policy ceilings, the Risk Committee sets those ceilings, and token holders progressively assume the authority the team delegates.

Security & trust assumptions

What you must trust, and what the protocol enforces.