Control model
An autonomous agent may propose a request, but that request should not automatically imply unlimited authority. MAGNIFIC inserts a policy layer between intent and execution so approvals become explicit rather than assumed.
System layers
Produces the target, asset, amount and requested action that enters the control path.
Applies explicit rules such as allowlists, ceilings and action constraints.
Receives only those requests that satisfy the active policy decision path.
Stores the decision context and resulting outcome for later review.
Policy primitives
The operating model is designed around simple policy primitives that remain readable for both developers and operators.
- Asset allowlists
- Target contract allowlists
- Maximum transaction value
- Daily operating budget
- Allowed action or function scope
- Pre-execution simulation requirements
Execution path
When a request satisfies the active policy, it becomes eligible for a connected execution environment. The website interface presents the decision model, while execution can remain separate from the policy surface.
Reviewable records
Each decision can be associated with execution context such as request identifiers, target contracts, values, timestamps and resulting status. The goal is simple: make autonomous behaviour easier to inspect after the fact.