Cómo usar Claude mejor que el 99%
This tutorial explains how to use Claude beyond basic chatbot interactions by leveraging its full ecosystem of features. The speaker covers prompt construction, file uploads, web search, and advanced features like Skills, Connectors (MCPs), Projects, and Artifacts. The goal is to transform Claude from a simple Q&A tool into a structured, reusable work system.
Summary
The video opens by arguing that most users treat Claude like a basic chatbot — asking a question and reading the answer — when in reality it functions as a full work system with memory, reusable processes, external integrations, and visual output capabilities. The speaker promises to cover both fundamentals and advanced features.
On the basics, the speaker explains the Claude interface, noting the sidebar, model selector, and file attachment options. He advises that the free tier is sufficient for casual use, but that a paid plan makes sense once Claude becomes part of a regular workflow. Regarding model selection, he recommends Sonnet for everyday balanced tasks (especially with extended thinking enabled) and Opus for complex, demanding work, while noting Opus is slower and more resource-intensive.
A central section focuses on prompt quality. The speaker argues that poor results are almost always the user's fault, not Claude's — vague prompts produce vague answers. He recommends structuring prompts around three elements: the instruction (what to do), the context (who you are, what situation you're in, what constraints apply), and the conditions (tone, format, length, things to avoid). He also suggests adding a meta-instruction at the end of prompts asking Claude to identify missing context or ambiguities before responding, which consistently improves output quality.
The speaker then covers file uploads (PDFs, images, CSVs, spreadsheets), showing how uploading real data — like YouTube analytics or thumbnail screenshots — shifts Claude from giving generic advice to performing specific, actionable analysis. He notes Claude can analyze images but cannot yet generate them from scratch.
On web search, he recommends a two-step approach: first ask Claude to research and contextualize a topic, then make the actual request. This separation of 'search' and 'execute' phases significantly improves accuracy on time-sensitive topics.
The advanced section covers four major features. Skills are reusable instruction sets that encapsulate a process, tone, and framework for a recurring task — for example, a hook-writing skill for YouTube videos. The speaker demonstrates creating one via Claude's built-in 'Skill Creator' and shows a side-by-side comparison proving skill-guided responses are more refined and consistent. Connectors (MCPs) allow Claude to integrate with external tools like Gmail, Google Calendar, Notion, Google Drive, and GitHub, turning it into a workflow command center rather than an isolated chat. Artifacts enable Claude to produce structured visual outputs — landing pages, dashboards, diagrams, simulators — directly within the chat interface, which can then be iteratively refined. Projects are persistent workspaces that maintain context, files, instructions, and conversation history across sessions, eliminating the need to re-explain background information for recurring work types. The speaker clarifies that Projects provide stable context while Skills refine execution, and combining both unlocks Claude's full potential. The video closes by mentioning Claude's desktop application, which also includes Cowork and a coding-focused interface.
Key Insights
- The speaker argues that poor Claude outputs are almost never the tool's fault — vague prompts produce vague answers, and structuring prompts around three components (instruction, context, and conditions) consistently improves response quality.
- The speaker demonstrates that adding a meta-instruction at the end of a prompt — asking Claude to identify missing context or ambiguities before responding — causes Claude to validate its understanding first, which makes final answers more specific and useful.
- The speaker shows that Skills are reusable instruction sets that encapsulate a process, tone, and framework, and demonstrates with a side-by-side comparison that a skill-guided hook ('AI doesn't do the job better because it's smarter; it does it better because it doesn't get tired') is noticeably more refined than a generic one.
- The speaker recommends a two-step web search workflow — first asking Claude to research and contextualize a topic, then making the actual request — arguing this separation produces more accurate, less generic answers on time-sensitive subjects.
- The speaker draws a key distinction between Projects and Skills: Projects provide stable, persistent context (files, instructions, conversation history) for recurring work types, while Skills refine the execution of specific tasks — and combining both is what makes Claude behave like a serious work system rather than a chat tool.
Topics
Transcript
[0:00] Cloud remains one of the most powerful tools you can use right now, but most people are still using it as if it were just another chatbot. They log in, write a question, receive an answer, and that's it. And sure, that's how it works, yes, but you're only scratching the surface because when you understand how to use the cloud properly, it stops being just a chat and becomes a work system. A system with projects that remember context, skills that reuse processes, connectors or [0:31] MCPs to bring real information from other tools, artifacts to build things within the cloud, and a much smarter way of working with prompts, documents, and research. And that's exactly what we're…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Migue Baena IA
Cómo aprender por tu cuenta y volverte imparable con IA
This tutorial demonstrates how to use Gemini Notebook (formerly Notebook LM) to self-teach any skill by curating personal learning sources and converting them into multiple study formats. The four-step process involves selecting quality materials, studying them through different media, teaching the content to others, and integrating learning into daily routines.
Tienes que probar esto antes de crear tus videos con IA 🎬
A new AI video creation tool called 3D Builder within Topw Film Studio allows creators to pre-visualize and plan movie scenes by building interactive 3D environments, positioning characters and objects, controlling camera movements, and reviewing animations before final video generation.
No elijas entre Claude y Codex: úsalos juntos 🤝🏻
The video proposes using Claude and Codex together rather than choosing one over the other to maximize their respective strengths. Claude handles problem analysis and task delegation while Codex executes code generation, allowing users to work around individual API rate limits and leverage complementary capabilities.
Cómo usar Claude mejor que el 99% de la gente
A comprehensive guide to using Claude AI effectively, covering account setup with context import, model selection and reasoning levels, proper prompting techniques, projects and skills for organizing work, and advanced features like Cowork for delegated tasks and Code for building applications.
El archivo de GitHub que deberías de conocer 👀
Anthropic has published a detailed GitHub repository containing a 275,000+ character PRD with nearly 8,000 lines of instructions for one of their models, including specifications on what to do, avoid, and how to use tools. The key value lies not in copying the content but in studying its structure and learning how it organizes rules and defines tool usage for prompt engineering applications.