# Scott Logic This feed is brought to you by Scott Logic's team of technical authors and bloggers, covering topics including HTML5, iOS, C#, process, UX and practically anything else relating to software development. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## MCP Apps: Your UI, Their Chat DevFeed: [MCP Apps: Your UI, Their Chat]() Original publisher: [Read original article]() Author: James Strong Published: 2026-09-16T07:00:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [Model Context Protocol (MCP)](), [ui](), [ChatGPT](), [App]() Tags: [ai](), [app](), [artificial-intelligence](), [chatgpt](), [mcp](), [ui]() ### AI overview The article examines MCP Apps, an extension of the Model Context Protocol, alongside the protocol's 2026-07-28 changes. It explains how MCP Apps let servers return isolated, interactive, stateful UIs that clients such as ChatGPT can embed in agent responses, with examples including dashboards, data tables, and forms. ### Source excerpt A lot of focus has been put on the statelessness introduced as part of the new MCP specification, but the formalisation of extensions, including MCP Apps, is also worth attention. Using FastMCP and Prefab, I put together an MCP app and show it running in ChatGPT. ## Why agentic AI starts with legacy modernisation DevFeed: [Why agentic AI starts with legacy modernisation]() Original publisher: [Read original article]() Author: Suzanne Angell Published: 2026-08-21T09:33:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [AI Agent](), [legacy](), [legacy systems](), [data](), [Processes]() Tags: [agentic-ai](), [agents](), [ai-agents](), [artificial-intelligence](), [data](), [governance](), [integration](), [legacy](), [legacy-modernisation](), [operational-risk](), [processes](), [risk-management]() ### AI overview The article argues that organisations must address fragmented data, brittle processes, and outdated architectures before they can realise the full value of agentic AI. It explains that legacy systems may remain operationally critical, while accumulated technical, process, and organisational debt can prevent agents from accessing reliable information and executing workflows consistently. ### Source excerpt Organisations are increasingly excited by the potential of agentic AI, but many overlook the legacy obstacles that stand in the way. In this post, I explore why successful AI adoption depends on tackling fragmented data, brittle processes and outdated architectures, and why modernisation is often the most important step towards unlocking value from intelligent agents. ## Want to use AI agents safely? Start with design DevFeed: [Want to use AI agents safely? Start with design]() Original publisher: [Read original article]() Author: Colin Eberhardt Published: 2026-08-18T13:12:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [AI Strategy](), [Security](), [Monitoring](), [Architecture & Design](), [Resilience]() Tags: [agentic-ai](), [agents](), [ai](), [artificial-intelligence](), [auditability](), [design](), [end-to-end-process](), [featured](), [governance](), [guardrails](), [monitoring](), [observability](), [operational-resilience](), [quality](), [risk-management](), [security](), [service-design](), [systems](), [trust]() ### AI overview This article argues that organisations adopting AI agents should begin with process and system design rather than controls alone. It explains that design should account for human and machine strengths, establish proportionate guardrails, and define how observability and monitoring evolve as the system matures. ### Source excerpt Concerns about control are one of the biggest barriers to adopting agentic AI, particularly in regulated environments. In this post, we discuss how organisations can harness AI safely by designing processes around the strengths of both humans and machines, then applying the right controls, guardrails and monitoring. ## How to take incremental steps towards data democratisation DevFeed: [How to take incremental steps towards data democratisation]() Original publisher: [Read original article]() Author: Andy Scotland Published: 2026-08-17T13:09:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [data](), [trust](), [decision-making](), [Artificial Intelligence]() Tags: [agent-factory](), [ai](), [ai-agents](), [artificial-intelligence](), [compliance](), [conway-s-law](), [data](), [data-democratisation](), [data-engineering](), [data-platform](), [data-products](), [governance](), [incremental](), [trust]() ### AI overview This article explains how organisations can pursue data democratisation incrementally while preserving control, risk management and governance. It describes potential benefits in financial services and discusses how trusted data, data products, platforms and AI agents could support innovation, decision-making and customer service. ### Source excerpt Organisations increasingly recognise the value of making data more accessible, but concerns around control, risk and governance often stand in the way. In this post, I explore why data democratisation doesn't require organisations to sacrifice oversight, and how data products, platforms and agent factories can unlock innovation while maintaining trust, compliance and accountability. ## AI success depends on data foundations DevFeed: [AI success depends on data foundations]() Original publisher: [Read original article]() Author: James Heward Published: 2026-08-14T14:19:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [data-architecture](), [AI Strategy](), [AI Development]() Tags: [agentic-ai](), [ai](), [ai-readiness](), [artificial-intelligence](), [data](), [data-engineering](), [governance](), [integration](), [legacy-it](), [legacy-modernisation](), [prototypes](), [quality](), [silos]() ### AI overview Strong data foundations--discoverable, high-quality, accessible, governed, and integrated--are presented as essential for reliable AI outcomes, especially as organisations adopt more autonomous agentic systems. ### Source excerpt As organisations invest in AI, many discover that their biggest challenges are not AI-related at all. In this post, I explore why strong data foundations, from quality and accessibility to governance and integration, are essential for turning AI ambition into reliable, production-ready outcomes. ## Unlocking your data: the value is in collaboration DevFeed: [Unlocking your data: the value is in collaboration]() Original publisher: [Read original article]() Author: Sam Perridge Published: 2026-08-12T14:59:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [data](), [datasets](), [Artificial Intelligence](), [decision-making](), [systems]() Tags: [ai](), [business-intelligence](), [claude](), [collaboration](), [data](), [data-democratisation](), [data-engineering](), [data-maturity](), [data-platform](), [datasets](), [decision-making](), [governance](), [guardrails](), [reporting](), [self-service]() ### AI overview This opinion article argues that organisations unlock more value from data when datasets are connected and insights are accessible across teams. It describes a progression from paper records and siloed systems to connected and democratised data, including self-service analytics and AI, while emphasising governance and practical adoption. ### Source excerpt Organisations often focus on collecting data and connecting systems, but the greatest value comes from helping datasets work together and making insights accessible to the people who need them. In this post, I explore the journey from siloed data to democratised access, showing how self-service analytics and AI can unlock hidden value, while strong governance provides the guardrails for confident decision-making. ## Agile Leadership DevFeed: [Agile Leadership]() Original publisher: [Read original article]() Author: Dave Ogle Published: 2026-08-10T00:00:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Agile](), [Self-organizing Team](), [trust](), [Development]() Tags: [agile](), [delegation](), [delivery](), [frameworks](), [leadership](), [self-organizing-teams](), [software-development](), [teams](), [trust]() ### AI overview This opinion argues that organisational agility depends on leadership as well as agile frameworks and processes. It highlights empowerment, trust and delegation, and examines Agile2 and the British Army's Mission Command philosophy as relevant perspectives. ### Source excerpt Agile teams are built on more than frameworks and processes. This article explores why effective leadership, trust and delegation are fundamental to true organisational agility, drawing lessons from the British Army's Mission Command philosophy. ## How to accelerate agentic adoption DevFeed: [How to accelerate agentic adoption]() Original publisher: [Read original article]() Author: Simon Sear Published: 2026-08-06T08:37:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [Software Engineering](), [Automation]() Tags: [adoption](), [agentic-ai](), [agents](), [ai](), [ai-adoption-spectrum](), [apps](), [artificial-intelligence](), [automation](), [business-process](), [claude](), [delivery](), [engineering](), [safely](), [software-engineering]() ### AI overview This article presents a framework for accelerating agentic AI adoption by grouping initiatives into four categories: employee productivity tools, business transformation and workflow automation, AI features embedded in existing applications, and AI-enabled software engineering. It emphasizes clear direction, safe deployment, secure data access, education, and organizational change. ### Source excerpt Agentic AI is impacting every part of an organisation, from customer journeys to back office, supply chains and software engineering. Everything is changing at an accelerating speed, all at the same time, everywhere, all at once. In response, organisations don't want drawn-out analysis engagements with consultants or long strategy projects; they want action. They want to quickly and safely deploy agents to production and deliver value fast. ## AI-Assisted Techniques for Migrating a Legacy Java Codebase to Rust DevFeed: [AI-Assisted Techniques for Migrating a Legacy Java Codebase to Rust]() Original publisher: [Read original article]() Author: pedwin@scottlogic.com (Paul Edwin) Published: 2026-08-04T00:00:00Z Content type: tutorial Language: en Sources: [Scott Logic]() Topics: [migration](), [Artificial Intelligence](), [Rust](), [AI-assisted coding](), [developer tooling](), [test]() Tags: [agents](), [ai](), [artificial-intelligence](), [code](), [compiler](), [migration](), [rust](), [test](), [testing]() ### AI overview The article discusses using AI developer tooling to help migrate a complex, under-tested legacy Java codebase to Rust. It focuses on maintaining confidence in generated code and tests, using Rust's compiler feedback and tooling to guide AI-assisted development. ### Source excerpt Migrating a legacy codebase of any reasonable complexity has always been challenging, but AI tools have fundamentally changed the economics of the process. Generating code is now cheap and fast, but we still need confidence that the generated code is as good as, or better than, code we'd write ourselves. ## Four pillars of agentic AI success DevFeed: [Four pillars of agentic AI success]() Original publisher: [Read original article]() Author: Simon Sear Published: 2026-07-29T15:47:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [Software Engineering](), [data](), [legacy](), [systems](), [context]() Tags: [accessibility](), [agentic-ai](), [ai](), [ai-accelerated-development](), [ai-agents](), [architecture](), [artificial-intelligence](), [context](), [data](), [data-architecture](), [data-governance](), [data-strategy](), [delivery](), [engineering](), [governance](), [legacy](), [legacy-modernisation](), [software-engineering](), [systems]() ### AI overview This article argues that organizations need strong foundations before scaling agentic AI. It identifies four pillars: modernizing legacy systems, building reliable data foundations and architecture, managing governance and risk, and democratizing access to data. ### Source excerpt Everyone is talking about agentic AI. That's not surprising. The promise is huge: AI agents that can plan, reason, use tools, work across systems and get real work done. In software engineering, that could mean faster delivery and better quality. In operations, it could mean complex processes moving with less manual effort, fewer handovers and better decisions. ## Shifting readiness left: What AI can (and can't) do for organisational readiness DevFeed: [Shifting readiness left: What AI can (and can't) do for organisational readiness]() Original publisher: [Read original article]() Author: Nel Mathams Published: 2026-07-27T14:48:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [data]() Tags: [ai](), [artificial-intelligence](), [building-shared-understanding](), [delivery](), [discovery](), [ethics](), [funding](), [governance](), [institute-for-government](), [objectives](), [organisational-goals](), [organisational-readiness](), [shift-left](), [transformation]() ### AI overview This opinion article argues that organisational readiness depends on shared understanding rather than only governance, funding, or delivery structures. It examines how AI can help align goals, surface hidden assumptions, and accelerate discovery while keeping human judgement in control, including ethical considerations around data. ### Source excerpt Organisational readiness is often treated as a matter of governance, funding and delivery structures. In this post, I argue that true readiness is about building shared understanding, and explore how AI can help organisations align on goals, surface hidden assumptions and accelerate discovery work, while keeping human judgement firmly in control. ## The Rise of Open-Weight AI Models and the Challenge to Commercial Frontier Labs DevFeed: [The Rise of Open-Weight AI Models and the Challenge to Commercial Frontier Labs]() Original publisher: [Read original article]() Author: Colin Eberhardt Published: 2026-07-27T00:00:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [AI Models](), [Open Source](), [OpenAI](), [anthropic](), [deepseek](), [ChatGPT](), [llama](), [Reinforcement learning](), [Transformer]() Tags: [2026](), [ai](), [ai-models](), [anthropic](), [artificial-intelligence](), [chatgpt](), [china](), [cost](), [deepseek](), [llama](), [open-source](), [openai](), [performance](), [privacy]() ### AI overview This article traces the development of commercial and open-weight AI models from OpenAI's founding through the rise of models predominantly emanating from China. It argues that open-weight models have narrowed or closed the capability gap with frontier labs and may offer comparable or better performance at lower cost, with greater ownership and data privacy. ### Source excerpt Chinese-led open-weight AI models have closed the capability gap with frontier labs like OpenAI and Anthropic, challenging the assumption that massive investments guarantee lasting competitive advantage. By 2026, open-weight alternatives offer equivalent or superior performance at a fraction of the cost, along with greater ownership and data privacy, fundamentally disrupting the commercial AI landscape. ## Choosing the right tool safety approach for coding agents DevFeed: [Choosing the right tool safety approach for coding agents]() Original publisher: [Read original article]() Author: Robat Williams Published: 2026-07-15T12:00:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [AI-assisted coding](), [developer tooling](), [prompt injection](), [AI Development](), [Claude Code]() Tags: [artificial-intelligence](), [build](), [claude-code](), [code](), [coding-agents](), [prompt-injection](), [safety]() ### AI overview This article examines safety risks in coding agents, focusing on how models, runtimes such as Claude Code, and tools enable agents to act on their environment. It discusses permission-based harnesses, permission fatigue, prompt injection, and the limitations of relying solely on model alignment. ### Source excerpt In coding agents, the trio of: models, runtimes (e.g. Claude Code), and tools (e.g. file write) are what makes agentic software development possible. While they work safely most of the time, the (unsafe) default setup many of us fall into does carry risk. In this post, I'll explore some things you could consider to make your own setup safer - or at least help you understand its limitations. ## Adopting Agentic: Talks and Reflections from Our AI Engineering Event DevFeed: [Adopting Agentic: Talks and Reflections from Our AI Engineering Event]() Original publisher: [Read original article]() Author: Colin Eberhardt Published: 2026-07-15T00:00:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [AI Engineering](), [Software Engineering](), [Automation]() Tags: [agentic](), [ai](), [ai-engineering](), [automation](), [autonomous](), [engineering](), [event](), [reflections](), [sdlc](), [software-engineering](), [talks](), [tech]() ### AI overview This article summarizes talks and discussions from the Adopting Agentic event, focusing on how AI is affecting software engineering, individual skills and job roles, teams, organizational practices, and the software development life cycle. It also includes the author's reflections, key quotes, and session videos. ### Source excerpt A summary of the talks and discussions from our recent event, Adopting Agentic - Software Engineering for the AI Age, exploring the human and organisational impact of AI on our industry. Includes session videos, key quotes and my personal reflections on each presentation. ## GLM-5.2: Considerations for enterprise teams starting out with open-weight models DevFeed: [GLM-5.2: Considerations for enterprise teams starting out with open-weight models]() Original publisher: [Read original article]() Author: Robat Williams Published: 2026-07-08T10:00:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [Agentic development](), [developer setup](), [AI Development](), [experiments](), [Benchmark](), [Open Source](), [Claude]() Tags: [agent](), [agentic](), [agentic-development](), [artificial-intelligence](), [benchmark](), [claude](), [developer-setup](), [experiments](), [open-source]() ### AI overview The article describes how a small team prepared an open-weight agentic development setup for AI productivity experiments. It explains the team's model-selection criteria, distinguishes open-weight models from locally run models, and reports choosing GLM-5.2 after comparing DeepSeek V4, Kimi K-2.6, and other models using benchmarks and practical trials. ### Source excerpt Setting yourself up to try out open-weight models for agentic development isn't difficult, but it isn't as straightforward as downloading a coding agent from one of the handful of well-known AI vendors. In preparation for the latest round of our AI productivity experiments, we've recently been through this process. Read on for the choices we made, the considerations at play, and what made our situation unusual. ## Auditable Agentic Orchestration: From Autonomous Systems to Governed Execution DevFeed: [Auditable Agentic Orchestration: From Autonomous Systems to Governed Execution]() Original publisher: [Read original article]() Author: Tom Stavert Published: 2026-07-08T00:00:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [agent orchestration](), [AI Agent](), [Orchestration](), [workflow automation](), [systems](), [Automation]() Tags: [agentic](), [ai-agents](), [artificial-intelligence](), [auditability](), [automation](), [distributed-systems](), [orchestration](), [workflow]() ### AI overview This article explains how agentic orchestration combines the flexibility of autonomous AI agents with deterministic workflow structures, explicit guardrails, and validation. It discusses workflow engines and agentic subprocesses as mechanisms for improving auditability and control without eliminating adaptive agent behavior. ### Source excerpt Giving AI agents free rein is easy. Trusting their output is hard. This post explores how workflow engines like Fluxnova, and its new Agentic Subprocess, bring auditability and control to agentic orchestration without sacrificing the flexibility that makes agents valuable. ## Agentrification and the Agentrification Index DevFeed: [Agentrification and the Agentrification Index]() Original publisher: [Read original article]() Author: Graham Odds Published: 2026-06-29T00:00:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [digital]() Tags: [agentic-ai](), [ai](), [ai-agents](), [artificial-intelligence](), [autonomous](), [autonomous-agents]() ### AI overview This satirical post defines "Agentrification" as the displacement of human character and participation in spaces, platforms, and professions by autonomous AI agents. It introduces the Agentrification Index (AIx) as a diagnostic tool for quantifying this structural decay. ### Source excerpt This satirical post defines Agentrification as the sociotechnical phenomenon in which the original character of a human-centric space, platform, or profession is systematically displaced, optimised, or rendered obsolete by the arrival of autonomous AI agents, i.e. digital gentrification. It introduces the Agentrification Index (AIx) as a diagnostic tool to quantify this structural decay. ## Sustainable acceleration and the Agentic Software Development Life Cycle DevFeed: [Sustainable acceleration and the Agentic Software Development Life Cycle]() Original publisher: [Read original article]() Author: Dan Allsop Published: 2026-06-19T00:00:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [sdlc](), [AI Development](), [AI Agent](), [Software Engineering](), [Prompt Engineering](), [engineering-culture]() Tags: [agentic](), [agentic-ai](), [agile](), [ai](), [ai-agents](), [artificial-intelligence](), [asdlc](), [cognitive-entropy](), [culture](), [delivery](), [governance](), [people](), [prompt-engineering](), [technical]() ### AI overview The article discusses the transition from the traditional Software Development Life Cycle to an Agentic Software Development Life Cycle, in which AI agents participate across requirements, design, implementation, verification, and maintenance. It argues that sustainable productivity gains require governance, clear policies, human oversight, disciplined engineering practices, training, architectural clarity, and prompt engineering. ### Source excerpt A discussion of how to successfully leverage agentic AI to raise the productivity ceiling covering the topics of human oversight, governance, and disciplined engineering practices that preserve stability. ## Working Effectively with Claude Code DevFeed: [Working Effectively with Claude Code]() Original publisher: [Read original article]() Author: Amy Laws Published: 2026-06-18T09:09:00Z Content type: article Language: en Sources: [Scott Logic]() Topics: [Claude Code](), [parallel](), [Terminal](), [Code](), [Orchestration](), [Git]() Tags: [agent](), [agents](), [artificial-intelligence](), [claude-code](), [code](), [copilot](), [github-copilot](), [orchestration](), [parallel](), [round-robin](), [terminal](), [vs-code]() ### AI overview This article shares practical lessons from switching from GitHub Copilot in VS Code to Claude Code. It focuses on parallel development with multiple agents, using terminal sessions and Git worktrees, and managing concurrent work through Claude Code's agent view. ### Source excerpt After months working with GitHub Copilot in VS Code and recently switching to Claude Code, the transition turned out to be more involved than expected. Claude Code operates differently and in ways that take time to adjust to. In this post, I share my experiences and tips drawn from that experience. ## Ponytail's benchmarked benefits may largely reflect simple YAGNI-style prompting DevFeed: [Ponytail's benchmarked benefits may largely reflect simple YAGNI-style prompting]() Original publisher: [Read original article]() Author: Colin Eberhardt Published: 2026-06-16T16:27:00Z Content type: opinion Language: en Sources: [Scott Logic]() Topics: [Artificial Intelligence](), [prompt](), [AI-assisted coding](), [AI Agent]() Tags: [ai](), [ai-coding](), [artificial-intelligence](), [code](), [evaluation](), [prompt]() ### AI overview The article examines Ponytail, a prompt-based skill for AI coding agents that aims to reduce over-engineering. It reports that short prompts invoking YAGNI principles matched or exceeded Ponytail's benchmark results, and argues that the tool's attention was not supported by sufficiently robust evidence. It also notes that the author later expanded the benchmarks and revised the claims in response to the criticism. ### Source excerpt The post examines Ponytail, a popular AI coding "skill", and argues that its benchmarked benefits appear to come largely from encouraging terse, YAGNI-style responses rather than from any deeper engineering value. By showing that a simple prompt can match or beat Ponytail on its own benchmark, it makes a broader case for treating prompt-based tools with scepticism unless their claims are backed by robust evaluation.