This paper Metacybernetics: Aspect Traits and Fractal Patterns in Higher‑Order Cybernetics develops a formally grounded account of recursive agency by deriving the architecture of metacybernetics directly from first principles in information physics. Building on the metacybernetic framework introduced in Yolles (2021), the paper demonstrates that living and intelligent systems must follow a strict recursive pattern generated by the third‑order Potential/Actuation dyad. Using Fisher Information Field Theory (FIFT) within an Informational Realism paradigm, the paper derives the canonical 2–3–2–3–2 trait alternation that characterises fifth‑order cybernetic systems, the characteristion being represneted by 36 traits, each with two epistemically independent poles, giving 72 descriptors. This cannonical trait alternation emerges as a necessary consequence of informational parsimony, a functional analogue of the holographic principle, and prevents combinatorial explosion as systemic depth increases. In principle this can be extended nth-order cybernetic systems. The analysis shows how the R(3) sustentative system functions as a fractal generator whose structure propagates across higher cybernetic orders, shaping the organisation of cognition, affect, and conation in Mindset Agency Theory. The paper formalises the closure condition of the seven‑trait viability schema, explains its inheritance across aspects and recursive orders, and demonstrates how this fractal kernel underpins the emergence of higher‑order process intelligences. The Cogitor5 model is presented as a complete fifth‑order exemplar, illustrating how recursive coupling, holographic encoding, and dyadic–triadic parity enable self‑evolution, systemic coherence, and collective intelligence in biological, organisational, and artificial systems. The paper has two parts. Part I revisits the metacybernetic hierarchy and situates it within the broader traditions of cybernetics, autopoiesis, enaction, and complexity science. Part II provides the formal derivation of the recursive architecture using variational analysis on an implicate–explicate manifold, establishing the informational boundary conditions that generate fractal self‑similarity across scales. The paper concludes by outlining implications for intelligent system design, trait‑based diagnostics, and future empirical validation. This preprint contributes a rigorous, information‑theoretic foundation for higher‑order cybernetics, offering a generative framework for understanding recursive agency, trait inheritance, and the emergence of coherence in complex adaptive systems.