Pre-execution authority
Foundational Doctrine for Autonomous Systems
Version 1.0
February 2026
Abstract
Pre-Execution Authority establishes a foundational architectural principle:
No autonomous entity with decision-making capacity should execute actions without an explicitly declared authority framework.
This document defines the structural legitimacy required before execution.
It does not regulate how systems execute.
It defines the structural legitimacy required before execution.
1. The Structural Gap
Modern autonomous systems focus primarily on:
- Capability
- Orchestration
- Tool access
- Memory
- Runtime execution
- Performance optimization
However, a structural gap remains:
Autonomy is being engineered without explicit pre-declared authority.
Identity alone does not define limits.
Execution capability does not define legitimacy.
Payment capability does not define authorization.
The absence of declared authority creates architectural ambiguity.
2. Core Principle
The foundational principle of Pre-Execution Authority is:
Autonomy must be preceded by declared authority.
Execution without declared authority introduces systemic risk, governance ambiguity, and irreversible operational drift.
Authority must be:
- Explicit
- Structured
- Versioned
- Revocable
- Auditable
Before execution occurs.
3. Definition
Pre-Execution Authority is the declarative authority framework that defines:
- Operational legitimacy
- Autonomy scope
- Decision thresholds
- Governance boundaries
- Escalation conditions
- Revocation mechanics
It exists prior to runtime execution.
It is not an enforcement mechanism.
It is not a compliance regime.
It is not a runtime protocol.
It is a prerequisite structural declaration.
4. Structural Components
Pre-Execution Authority may be expressed through six conceptual blocks:
4.1 Origin Declaration
- Who created the autonomous entity
- Under which organizational or institutional authority it operates
4.2 Purpose Scope
- Declared domain of operation
- Explicit operational boundaries
4.3 Capability Envelope
- What actions are permitted
- What actions are explicitly prohibited
4.4 Autonomy Gradient
A declared autonomy level, aligned with the normative interpretation in spec v1.0:
- Level 0 — No execution: advisory / descriptive only
- Level 1 — Supervised: actions only under explicit user instruction or approval
- Level 2 — Scoped autonomy: bounded workflows within declared constraints
- Level 3 — High autonomy: execution within declared constraints without per-step approval
This document is non-normative. The authoritative definition of autonomy.level is the canonical specification (spec/v1.0/spec.md).
Autonomy becomes a declared variable, not an implicit assumption.
4.5 Escalation Map
- Defined pause conditions
- Human or institutional escalation triggers
- Threshold-based decision boundaries
4.6 Revocation & Audit Model
- Suspension mechanisms
- Versioning and traceability
- Structural inspection capability
5. Fractal Responsibility
Pre-Execution Authority applies across scales:
- Single agents
- Multi-agent systems
- Swarm architectures
- Agent marketplaces
- Machine payment systems
- Autonomous financial actors
Responsibility scales fractally with autonomy.
As execution scope increases, authority clarity must increase proportionally.
6. Universality
Pre-Execution Authority is architecture-agnostic.
It does not depend on:
- Specific runtimes
- Identity protocols (DID, ERC standards, etc.)
- Payment rails
- Orchestration frameworks
- AI model providers
It precedes them.
It defines legitimacy before capability.
7. Relationship to Agent Manifest
Agent Manifest provides a declarative instrument for expressing identity, purpose, and boundaries.
Pre-Execution Authority defines the philosophical and structural doctrine that justifies such declarations.
Agent Manifest is an implementation surface.
Pre-Execution Authority is the foundational principle.
This document does not modify the Agent Manifest specification.
8. Foundational Statement
Autonomous systems are increasingly capable of acting, transacting, and deciding within complex environments.
If autonomy expands without declared authority, systemic instability becomes increasingly likely.
Pre-Execution Authority asserts that:
Legitimacy must precede execution.
This is not a constraint on innovation.
It is a condition for sustainable autonomy.
Author
Hernán Alfredo Capucci
Independent Researcher
ORCID: 0009-0008-7216-3032