Feature Requests

I just heard about this awesome AI Agent Automation tool tool that also has a chrome extension that works in browser with a visual workflow builder (like make.com , n8n, zapier) to easily build agents 'MindStudio provides AI Agents that help users automate tasks. Users can access these agents on any website with a Chrome extension. The platform allows users to build custom agents without coding skills, utilizing language, image, and voice models. These AI Agents can be used across various industries and sizes of businesses for tasks such as content creation, customer support, data management, and community moderation. MindStudio offers a free trial and resources like tutorials and expert support to help users get started.' https://www.mindstudio.ai How awesome would it be to have Merlin do this too?

Is it not possible to remove all merlin ai keywords flagging? I think it's completely unnecessary to be honest, take emochi for example as they didn't implement any limit. Let the model itself withold the restrictions rather than merlin itself. I for one wanted to use grok 3 to it's full capabilities without Merlin flagging certain keywords

Dear MERLIN API Team, I am writing to express my deep appreciation for the MERLIN API and to request that you reconsider the decision to discontinue this incredible product. As a loyal user of MERLIN API, I have found it to be an invaluable tool that has greatly enhanced my productivity and efficiency. The ease of integration and the wide range of built-in AI capabilities have been instrumental in helping me achieve my goals. Your product has truly been a game-changer for me. I understand that discontinuing a product is never an easy decision, and there may be various factors that led to this choice. However, I strongly believe that MERLIN API has a unique value proposition that sets it apart from other offerings in the market. Its user-friendly interface and powerful features have made it an essential part of my workflow. While I respect your decision, I kindly request that you consider the impact this discontinuation will have on users like myself who have come to rely on MERLIN API. If there is any possibility of keeping the product available, even with limited support or updates, it would be greatly appreciated by the user community. If the discontinuation is inevitable, I would be grateful if you could provide clear guidance on alternative solutions or recommendations for products that can offer similar functionality. This would help ease the transition and ensure that users can continue to meet their needs. Once again, I want to express my heartfelt gratitude for the exceptional product you have created. MERLIN API has been a true asset, and I sincerely hope that you will reconsider its discontinuation. Thank you for your attention to this matter and for your commitment to delivering innovative solutions. I look forward to your response.

Dear Team,Could you please consider adding the Perplexity model with web vision to your roadmap? Its unique qualities and responsive nature make it a valuable addition.Thank you!

I think it would be awesome to have a "Prompt Enhancer" feature. It could use a clickable icon to take the user's prompt and refine it, which I believe would really improve LLM results and overall user experience. Using a more affordable model like Gemini 1.5 Flash or GPT 4o-mini could make this feasible and super useful.

Features Intelligent Code Completion Provides advanced, AI-powered code suggestions and autocompletion options tailored to the current context. This feature not only supports standard completion but also integrates deep learning models (similar to GitHub Copilot or TabNine) to deliver more sophisticated code generation. This reduces manual input and significantly boosts development efficiency. Advanced Error Detection and Fix Suggestions Performs real-time syntax checks and delivers AI-driven, context-aware recommendations for error detection and fixing. The plugin analyzes the broader scope of your code, identifies potential issues, and offers best-practice suggestions for resolution, surpassing traditional tools like ESLint or Pyright. Its accuracy and fine-grained analysis make debugging more intuitive. One-Click Documentation and Advanced Viewing Instantly access function signatures, class definitions, annotation descriptions, and other relevant documentation. While traditional tools provide static help, this plugin uses AI to dynamically generate context-appropriate suggestions. It enables natural language queries for targeted answers, such as "What parameters does this function accept?"—akin to the interaction style of embedded ChatGPT functionality. Intelligent Debugging Assistant Streamlines debugging tasks by automatically recommending actions for breakpoints, variable monitoring, and log output. Powered by AI analytics, the debugging assistant identifies potential causes of errors and suggests tailored solutions. This goes beyond VS Code's built-in debugging, providing an interactive experience comparable to PyCharm or IntelliJ IDEA's enhanced debugging tools. Project Indexing and Enhanced Code Navigation Automatically scans the entire project to create a comprehensive code index. The plugin leverages AI modeling to enable smarter symbol lookup, cross-file dependency analysis, and even natural language-based search queries for files, classes, or functions (similar to Kite's intelligent search). This boosts navigation speed and supports complex project visualization. Code Generation and Template Customization Includes built-in AI-powered code generation capabilities, enabling users to generate complete structures or templates from high-level descriptions. For example, it can produce a REST API or algorithm framework based on a task description, similar to OpenAI Codex or other intelligent coding tools. Additionally, users can customize frequently-used code templates for one-click use. Team Collaboration and Knowledge Sharing Equipped with a built-in knowledge-sharing function, this feature allows users to share specific code snippets or questions with teammates. The plugin further recommends relevant code snippets or best practices, mimicking JetBrains' Code With Me or collaborative document tools but focused on coding.

(PROOF OF DEMAND INSIDE) Having almost unlimited context so users can chat for months on end without facing the LLMs short context, you can add a sort of internal database, where with a button i can decide to store on your database thru embeddings or vector database, way longer context, this is useful and would make your clients stick with Merlin even if a new compeititor pops ups because the whole data, history and memory would be in your server. I also heard the hosts from "My first million" saying this and a lot of people replied with their own handmade solution that was unefficient and for developers, there's DEMAND FOR THIS: https://x.com/thesamparr/status/1869117964194124071?lang=en

I think you can ease a lot of uploading (I do it myself) if you have a desktop version which can read and perform action based on local files without the need to upload the whole file.

Would be nice if .ts (not only .tsx) files and .java files were uploadable.

I'd like to be able to choose models based on energy efficiency as one of my biggest concerns with AI is environmental impact. It'd be great if I could see the scoring next to each model. This could be a nice feature addition based on the below example from Hugging Face. https://huggingface.co/spaces/AIEnergyScore/Leaderboard