Developer Builds VS Code Tool to Map Live Call Chains After AI Agents Break Contracts

·6 views

A developer frustrated by AI coding agents silently breaking system contracts built Contour, a VS Code extension that automatically maps real-time call chains across Java/Spring Boot backends and React frontends. The tool addresses a common problem where agents or humans edit one side of a codebase without knowing downstream dependencies exist, such as scheduled jobs, Kafka events, or chained service calls. Contour reads source code directly to reflect how a system actually connects at any given moment, rather than relying on outdated documentation or diagrams. It surfaces inline CodeLens annotations in the editor gutter, showing which frontend calls hit which controller endpoints and vice versa, with explicit warnings when no caller or handler is found. A companion tool called Contour MCP was also built to let AI agents query the call-chain map directly, rather than operating with only single-file context.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Developer Builds Open-Source SDK to Track Brand Mentions Across AI Search Engines

A developer has released webscore-sdk, an npm package designed to help track brand visibility across AI platforms such as ChatGPT, Gemini, Perplexity, and Claude. The tool targets a practice known as Generative Engine Optimization (GEO), which focuses on improving how brands appear in AI-generated responses. The SDK measures two core metrics: mention rate, the percentage of relevant prompts where a brand appears, and citation rate, the percentage where the brand's website is linked. It also returns a sentiment reading and an overall GEO score ranging from 0 to 100. The package is available via npm, and a free tier offering 10 scans can be accessed at webscore.dev.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

12 Practical LINQ Examples Every .NET Developer Should Master

LINQ (Language Integrated Query), introduced in .NET 3.5, allows C# developers to query and manipulate collections using clean, expressive syntax instead of verbose loops. A DEV Community article outlines 12 practical LINQ examples covering key methods such as Where(), Select(), OrderBy(), and FirstOrDefault(). LINQ works consistently across arrays, lists, dictionaries, XML, and Entity Framework, making it a versatile tool in modern .NET development. The guide favors method syntax over query syntax, noting it is more concise and widely used in production codebases. The article also covers best practices, common mistakes, and interview questions to help developers strengthen their LINQ skills.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Millions of Job Openings Stay Hidden Because ATS Pages Evade Search Engines

A large share of open job positions never appear on mainstream job boards like LinkedIn or Indeed, despite being publicly accessible online. Most modern companies, particularly startups, manage hiring through third-party Applicant Tracking Systems such as Greenhouse, Lever, and Ashby, which host job listings on external pages rather than the company's own website. Because these ATS-hosted pages are often weakly linked or absent from sitemaps, search engine crawlers frequently fail to index them. The problem is compounded by companies regularly switching ATS providers, changing URLs, and updating listings throughout the day, making consistent aggregation difficult. The insight emerged from research conducted while building WorkWay, a job discovery platform designed to surface roles that traditional job searches routinely miss.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Anthropic's Constitutional AI gives Claude auditable ethics, setting it apart from rivals

Anthropic introduced Constitutional AI (CAI) in 2022, training its Claude model against an explicit set of written principles inspired by sources such as the Universal Declaration of Human Rights. Unlike competing models that rely heavily on human annotators for reinforcement learning, Claude uses a process called RLAIF, where the model critiques and revises its own responses based on those documented principles. This makes the model's decision-making transparent and auditable, a quality that carries practical value in regulated industries where explainability is a legal or compliance requirement. In comparative testing across contract analysis and smart-contract scenarios, Claude showed more consistent and explainable refusals when pushed toward generating harmful outputs, compared to GPT-4 and Gemini. Anthropic has reported that CAI measurably reduces harmful responses without significantly degrading the model's overall usefulness, addressing a trade-off that has historically undermined other AI systems.

0 comments Read more at DEV Community