pith. machine review for the scientific record. sign in

arxiv: 2605.07376 · v1 · submitted 2026-05-08 · 💻 cs.SE

Recognition: 2 theorem links

· Lean Theorem

Low-code and no-code with BESSER to create and deploy smart web applications

Iv\'an Alfonso , Armen Sulejmani , Aaron Conrardy , Jordi Cabot
Authors on Pith no claims yet

Pith reviewed 2026-05-11 01:43 UTC · model grok-4.3

classification 💻 cs.SE
keywords low-codeno-codesmart web applicationsAI agentsopen-source frameworkweb-based editorapplication generationdeployment
0
0 comments X

The pith

An open-source low-code framework lets users design, generate, and deploy smart web applications containing AI agents through a free web-based editor.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents BESSER as a low-code approach that models application components at a high level and uses automated generators to produce runnable web applications with embedded AI agents. Commercial low-code platforms often create risks of vendor lock-in and restricted customization, which the authors aim to sidestep by releasing the entire system as open source with a publicly accessible editor. If the method works as described, developers could build and modify intelligent web apps with far less manual coding while retaining full control over the generated code and deployment process.

Core claim

BESSER is an open-source low-code framework that allows users to design, generate and deploy their application via a freely accessible web-based editor, while guaranteeing transparency and extensibility.

What carries the argument

The BESSER web-based editor that combines high-level models of application components with automated code generators to produce deployable smart web applications.

If this is right

  • Developers gain the ability to create and modify applications containing AI agents without writing most of the underlying code.
  • Generated applications can be deployed directly from the editor without additional proprietary services.
  • Full access to source code and models removes the risk that changes in a commercial platform will break existing applications.
  • Community contributions can extend the set of available components and generators over time.
  • Transparency in the generation process makes it possible to audit and verify how AI agents are integrated into the final web application.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same modeling approach could be adapted to generate applications for other platforms beyond web, such as mobile or desktop, if generators are added.
  • Integration with existing open-source AI libraries might become simpler because the generated code remains fully editable and inspectable.
  • Wider adoption could create a shared library of reusable models for common smart-web patterns, reducing duplicated effort across projects.
  • Empirical measurements of development speed and maintenance cost compared with hand-coded or commercial alternatives would strengthen the case for the approach.

Load-bearing premise

That releasing the framework as open source with a web editor will effectively solve the drawbacks of commercial platforms such as vendor lock-in and limited extensibility.

What would settle it

A direct comparison study in which teams build the same smart web application once with BESSER and once with a commercial platform, measuring time to deployment, ability to customize beyond the editor, and any signs of lock-in when switching tools.

Figures

Figures reproduced from arXiv: 2605.07376 by Aaron Conrardy, Armen Sulejmani, Iv\'an Alfonso, Jordi Cabot.

Figure 1
Figure 1. Figure 1: BESSER pipeline for smart web applications [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: GUI editor in BESSER B-UML, inspired by IFML [2] and extended with constructs for rich view components (e.g., charts and chat interfaces) and styling properties (e.g., colors and layout). The GUI can link to the structural and agent models, consolidating all models into a smart web application. The generation pipeline orchestrates specialized sub-generators, each responsible for a distinct technology layer… view at source ↗
read the original abstract

The increasing demand for web applications containing AI-agents, seen as smart web applications, has prompted the need for new techniques to facilitate their creation. Low-code has risen as an approach that reduces the amount of handwritten code by focusing on the abstraction of components in the form of models combined with automated generators to produce applications. Existing low-code platforms are commercial, leading to drawbacks such as the risk of vendor lock-in, limited extensibility, and more. We present the open-source BESSER low-code framework, which allows users to design, generate and deploy their application via a freely accessible web-based editor, while guaranteeing transparency and extensibility.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The paper presents BESSER, an open-source low-code/no-code framework for designing, generating, and deploying smart web applications that incorporate AI agents. It uses a freely accessible web-based editor for modeling, with automated generators to produce the applications, positioning this as a solution to drawbacks of commercial platforms including vendor lock-in and limited extensibility, while claiming to guarantee transparency and extensibility through its open-source release.

Significance. If the framework's design and implementation deliver measurable transparency (e.g., inspectable and modifiable generators) and extensibility (e.g., documented plugin points for custom AI components) as asserted, BESSER could provide a useful open alternative in the low-code space for AI-enhanced web apps. This might encourage community contributions and reduce barriers compared to proprietary tools, but the lack of usage data, extension examples, or comparative evaluations limits the assessed impact to potential rather than demonstrated.

major comments (1)
  1. Abstract and central claims section: The assertion that BESSER 'guarantees transparency and extensibility' (and thereby overcomes vendor lock-in and limited extensibility) is load-bearing for the contribution but rests only on the open-source status and web editor without any concrete mechanisms, code snippets, metamodel modification examples, generator inspection details, or side-by-side comparisons to tools like Mendix or OutSystems. This leaves the key differentiator unsupported.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive comments. We address the concern about supporting our claims on transparency and extensibility by expanding the manuscript with concrete details and examples in the revision.

read point-by-point responses

  1. Referee: Abstract and central claims section: The assertion that BESSER 'guarantees transparency and extensibility' (and thereby overcomes vendor lock-in and limited extensibility) is load-bearing for the contribution but rests only on the open-source status and web editor without any concrete mechanisms, code snippets, metamodel modification examples, generator inspection details, or side-by-side comparisons to tools like Mendix or OutSystems. This leaves the key differentiator unsupported.

    Authors: We agree that the current presentation relies on the open-source release to underpin the claims and would benefit from explicit mechanisms to make the differentiators clearer. In the revised manuscript we will add: (1) code snippets illustrating how the generators can be inspected and modified; (2) a worked example of metamodel extension for custom AI components; (3) documentation of the plugin/extension points; and (4) a concise textual comparison with Mendix and OutSystems that highlights differences in source-code access and customization. These additions will directly support the load-bearing claims without altering the core contribution. revision: yes

Circularity Check

0 steps flagged

No circularity detected; paper is a framework presentation without derivations or self-referential reductions.

full rationale

The manuscript presents the BESSER open-source low-code framework and asserts that its web-based editor and open-source release guarantee transparency and extensibility. No equations, fitted parameters, predictions, or derivation chains exist. The central claim is the artifact itself rather than a computed result that reduces to its own inputs by construction. No self-citations, ansatzes, or uniqueness theorems are invoked in a load-bearing way. The paper's structure is descriptive and artifact-focused, making it self-contained against external benchmarks with no internal circular steps.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

The paper introduces a new software artifact rather than a theoretical result. No free parameters, mathematical axioms, or invented physical entities are present beyond the framework itself.

invented entities (1)
  • BESSER framework no independent evidence
    purpose: To enable low-code and no-code creation and deployment of smart web applications
    BESSER is the novel software tool introduced by the authors as the central contribution.

pith-pipeline@v0.9.0 · 5409 in / 1128 out tokens · 40844 ms · 2026-05-11T01:43:50.264848+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

8 extracted references · 8 canonical work pages

  1. [1]

    Building

    Alfonso, Iv. Building. International Conference on Business Process Modeling, Development and Support , pages=. 2024 , organization=

    work page 2024

  2. [2]

    Towards the Interoperability of Low-Code Platforms

    Alfonso, Iv \'a n and Conrardy, Aaron and Cabot, Jordi. Towards the Interoperability of Low-Code Platforms. Intelligent Information Systems. 2025

    work page 2025

  3. [3]

    International Conference on Conceptual Modeling , pages=

    Vibe Modeling: Challenges and Opportunities , author=. International Conference on Conceptual Modeling , pages=. 2025 , organization=

    work page 2025

  4. [4]

    2017 , howpublished =

    work page 2017

  5. [5]

    Expert Systems , volume =

    Benaddi, Lamya and Ouaddi, Charaf and Jakimi, Abdeslam and Chaibi, Hasna and Chehri, Abdellah and Jeon, Gwanggil and others , title =. Expert Systems , volume =. doi:https://doi.org/10.1111/exsy.70124 , year =

    work page doi:10.1111/exsy.70124

  6. [6]

    2014 , publisher=

    Interaction flow modeling language: Model-driven UI engineering of web and mobile apps with IFML , author=. 2014 , publisher=

    work page 2014

  7. [7]

    A UML-Based Methodology for Hypermedia Design

    Hennicker, Rolf and Koch, Nora. A UML-Based Methodology for Hypermedia Design. UML 2000 --- The Unified Modeling Language. 2000

    work page 2000

  8. [8]

    Applying the Oows Model-Driven Approach for Developing Web Applications

    Fons, Joan and Pelechano, Vicente and Pastor, Oscar and Valderas, Pedro and Torres, Victoria. Applying the Oows Model-Driven Approach for Developing Web Applications. The Internet Movie Database Case Study. Web Engineering: Modelling and Implementing Web Applications. 2008

    work page 2008