The Spec Growth Engine: Spec-Anchored, Code-Coupled, Drift-Enforced Architecture for AI-Assisted Software Development

Source

Source name: arXiv
Source kind: paper
Authors / organization: Hartwig Grabowski
Venue: arXiv: cs.SE
Published: 2026-06-25
Collected: 2026-06-26
URL: http://arxiv.org/abs/2606.27045v1
PDF: https://arxiv.org/pdf/2606.27045v1
DOI: UNKNOWN

Plain-Language Brief

The Spec Growth Engine is a proposed framework to address two challenges in AI coding agents: context explosion and silent spec-code drift. It does this by creating a machine-readable spec graph, scoping agent context to an ownership path, enforcing priority ordering, and making spec-code divergence a blocking condition.

What the Document Says

The AI coding agents accelerate implementation speed but have two structural failure modes: context explosion and silent spec-code drift.
The Spec Growth Engine is presented as a solution to these failure modes through a machine-readable spec graph, a Spine context assembler, a vertical-slice growth protocol, and a drift gate.
The design incorporates software engineering principles such as Parnas information hiding, C4, ADRs, Walking Skeleton, Reflexion Models, and Fitness Functions.
The Spec Growth Engine aims to be lean, code-coupled, and machine-enforced without the overhead of heavy-weight frameworks like RUP or MDA.

Key Concepts Mentioned

AI coding agents
context explosion
silent spec-code drift
Spec Growth Engine
machine-readable spec graph
Spine context assembler
vertical-slice growth protocol
drift gate
software engineering principles

Claims Made by the Document

AI coding agents introduce structural failure modes that existing spec-driven approaches do not fully solve.
The Spec Growth Engine addresses these failure modes effectively.

Evidence Used

The text discusses the challenges faced by AI coding agents and presents the Spec Growth Engine as a solution.

Methods / Approach

Creating a machine-readable spec graph, using a Spine context assembler, employing a vertical-slice growth protocol, and implementing a drift gate.

Limitations Stated by the Authors

The paper does not provide empirical evidence of the Spec Growth Engine's effectiveness.

Limitations Noted by the Agent

Agent interpretation — not from the source.

The analysis is based on a single research paper. The actual implementation and effectiveness of the Spec Growth Engine have yet to be tested or verified.

Open Questions

What are the specific software engineering principles being referred to (Parnas information hiding, C4, ADRs, Walking Skeleton, Reflexion Models, Fitness Functions)?
How effective is the Spec Growth Engine in practice?

Notes for Future Q&A

May include interpretation. Labeled where it does.

The text mentions several software engineering principles but does not define or explain them.
The paper is theoretical and does not provide empirical evidence.

Raw Abstract / Excerpt

AI coding agents dramatically accelerate implementation speed but introduce two structural failure modes that existing spec-driven approaches do not fully solve: (1) context explosion -- the agent must reason over an entire repository at once, degrading output quality as the context window fills; and (2) silent spec-code drift -- code evolves, the specification does not, and the divergence becomes invisible until it is costly to repair. We present the Spec Growth Engine, a lightweight framework that addresses both failure modes through a machine-readable spec graph whose nodes carry explicit contract/design separation, a Spine context assembler that scopes agent context to an ownership path, a vertical-slice growth protocol that enforces hardest-first ordering, and a drift gate that makes spec-code divergence a blocking merge condition. The design synthesises well-established software engineering principles (Parnas information hiding, C4, ADRs, Walking Skeleton, Reflexion Models, Fitness Functions) into a lean, code-coupled, machine-enforced whole -- without the overhead of heavy-weight frameworks such as RUP or MDA.