Agent Manifest Research

Papers

Agent Manifest Papers is an ongoing series examining declarative governance in autonomous systems.

The normative definition of Agent Manifest is exclusively contained in the canonical specification (v1.x series).

The Papers series is analytical and non-normative.

The series explores the architectural implications of undeclared authority, implicit mandate, and the necessity of declaration before execution.

These papers do not define implementation and do not modify the normative specification.
They examine the structural conditions under which autonomous systems can operate with declared legitimacy.


Published

Paper #01 — Declaration Before Execution

A structural examination of authority before execution in autonomous AI systems.
https://medium.com/@brfnrvf/agent-manifest-papers-01-declaration-before-execution-6d6e35bbf1b6

Paper #02 — The Problem of Implicit Authority

A structural analysis of undeclared mandate in autonomous AI systems.
https://medium.com/@brfnrvf/agent-manifest-papers-02-the-problem-of-implicit-authority-bbf33306387d

Paper #03 — Authority Is Not Inherent

On delegated mandate in autonomous AI systems. https://medium.com/@brfnrvf/authority-is-not-inherent-2b3061f402cd

These papers are hosted externally and are independent publications by the author. They are not normative components of the specification.

Paper #04 — The Pre-Execution Identity Gap

Formalizes the pre-execution identity gap and describes a minimal declarative layer (Agent Manifest) for pre-execution boundary definition.

DOI: https://doi.org/10.5281/zenodo.19432968

This paper defines the Pre-Execution Identity Gap and argues that declarative identity may be treated as an architectural layer for multi-agent systems.


Scope

The Papers series operates strictly at the conceptual layer.

It does not introduce runtime components.
It does not perform validation.
It does not enforce policy.
It does not alter the declaration schema.

It examines the principles underlying pre-execution authority and declarative boundaries in autonomous systems.


External citations

External academic and technical works citing Agent Manifest are tracked in CITATIONS.md.

Ongoing

Paper #05 — In development. No identifier is registered yet; this entry records that the work exists, not that it is published.