The fenbs blogAll articles
Page 69 of 99
- Cursor Global Rules: Set Defaults Across Every ProjectCursor lets you write rules that apply to all your projects, not just one repository. Here is how to use global rules well and what belongs there versus in the project. · 6 min read
- Cursor Rules in Codex: Translate and Check Their ScopeMove useful Cursor instructions into AGENTS.md, preserve the boundaries around each rule, and check what Codex actually reads before starting work. · 5 min read
- Customer Success AI Assistant: Briefings and HandoffsUse a customer success AI assistant to prepare account briefings, draft follow-ups, and make handoffs clearer. Keep customer commitments tied to verified information and human approval. · 7 min read
- How to Make ChatGPT Think Deeper About a DecisionAsk for explicit assumptions, meaningful alternatives, and evidence that could change the recommendation. Judge the analysis by what you can check, rather than how much reasoning it appears to display. · 5 min read
- How to Use AI to Clean Up GmailA staged AI cleanup turns an overloaded Gmail into a manageable inbox without the risk of bulk-deleting something important. The key is reversible steps, clear triage rules, and saving your criteria for next time. · 4 min read
- LLM Compliance: A Practical Review for Workplace UseReview the data, purpose, people, and controls behind a workplace LLM use case. Keep human approval clear and make the operating rules available to every assistant. · 5 min read
- Kiro MCP Server Setup: Add and Verify Tool ConnectionsHow to add MCP servers to Kiro, test each connection, and set up fenbs so Kiro and any other connected assistant start from the same project decisions and next steps. · 5 min read
- Prompts for Books: Plan, Draft, and Revise With AIReusable AI prompts for every stage of writing a book: planning, outlining, drafting chapters, checking continuity, and revising for clarity. · 6 min read
- Roam Research Alternatives for Notes and Shared AI ContextChoose a Roam Research alternative by what you need to preserve: linked thinking, readable files, shared research, or the work context your AI apps should reuse. · 8 min read
- System Prompt vs Master Prompt: Meaning, Scope, and ReuseA system prompt is a formal API role. A master prompt is an informal term for a large pasted block of instructions. The difference matters when you are trying to make your rules consistent and reusable. · 5 min read
- Why Does ChatGPT Use So Many Line Breaks?ChatGPT breaks text into short, single-sentence lines because its training associated that style with readability. Here is how to ask for compact paragraphs and save the rule for future sessions. · 5 min read
- AI Agents for Mechanical Engineering: Context and Design ReviewAI agents help mechanical engineers with specification drafts, reference research, and documentation under professional review. Tools like fenbs preserve design assumptions, source links, and revision decisions across every session. · 5 min read
- AI for Enterprise Architects: Research and Decision SupportHow enterprise architects can use AI to compare technology capabilities, analyze application portfolios, and draft recommendations, while keeping agreed principles and decision reasoning available to every session. · 7 min read
- AI Implementation Roadmap: From First Use Case to ReviewUse a staged roadmap to introduce AI into one real team workflow. Give every stage an owner, evidence to collect, and a clear reason to continue or stop. · 5 min read
- AI Notes With Pictures: Read Diagrams and Check the DetailsUse AI to make notes from permitted images, handwriting, and diagrams, verify the details against the source, and keep source references and open questions where the next session can continue from. · 5 min read
- AI Policy for Work: A Practical TemplateUse this adaptable AI policy to set approved tools, information-sharing limits, human review, and ownership. Then make the agreed rules available where people and assistants work. · 5 min read
- AI UI Design Generators: From Brief to Reviewed ScreensCompare AI UI generators by the output you can edit, the requirements they preserve, and the handoff they leave. Choose with a small prototype you can actually review. · 7 min read
- ChatGPT Custom Instructions Not Working? Find the CauseDiagnose the scope, wording, conflicts, and missing context behind an ignored instruction. Use a small repeatable test so each change addresses the actual failure. · 5 min read
- ChatGPT for Planning: Turn a Goal Into Reviewed Next StepsUse ChatGPT to compare practical options, expose assumptions, and draft a plan you can check. Keep the approved direction and current tasks ready for the next session. · 4 min read
- Claude Google Calendar Connector: Setup and UseYou can connect Google Calendar to Claude using a Google Calendar MCP server or through broader Google Workspace access. Here is how to check which route works for your setup and keep meeting briefs and agreed follow-up tasks organized. · 5 min read
- CLAUDE.md: Write Clear Project Instructions for Claude CodeA CLAUDE.md file gives Claude Code the project instructions you would otherwise repeat. Learn what to include, where it belongs, and how to keep it useful. · 5 min read
- Claude Skills for Unity: Development Review TemplatesReusable Claude skills for reviewing Unity scene changes, component behavior and play-mode results, with fenbs preserving design decisions and the current development plan. · 5 min read
- Codex CLI Scheduled Tasks: Options and LimitsCodex CLI does not schedule itself. You use the operating system scheduler or an external platform to run it on a timetable. Here is how each option works, what limits to expect, and how to track each run's outcome. · 5 min read
- Does Gemini Code? What It Can Do and What to CheckGemini can write, review and explain code across several interfaces, from the web app to the Gemini CLI in your terminal. Here is what each option covers and what to keep in mind before you ship. · 6 min read
- Claude Export: Save Chats and Prepare a Useful HandoffRequest your Claude conversation export, review what it contains, and deliberately save the decisions and next steps your next session needs. · 5 min read
- Google AI for PDFs: Ask Questions and Check the SourcesUse Google AI to work through a PDF one question at a time, then check each useful answer against the original. Keep the source and your research progress together for the next session. · 5 min read
- How to Access ChatGPT Memory and Review What It RemembersFind the memory view available to your account, review its contents, and distinguish it from chat history or project sources. Keep essential work facts in a record you can share and verify. · 5 min read
- How to Give Claude Context Without Overloading the ChatGive Claude the goal, relevant evidence, settled decisions and current state in one compact starting note. Then check what it understood before asking it to do the work. · 5 min read
- How to Make Claude Use YouTube Videos in ResearchClaude cannot stream a YouTube video, but it can work with the transcript — and for most research tasks, the transcript is what matters. Here is how to get the content into Claude and how to verify what it finds. · 4 min read
- How to Read a PDF With AI and Check the AnswersUse AI to find the useful parts of a PDF, ask focused questions, and check the evidence on the page before you reuse an answer. · 5 min read
- How to Make Claude More Critical of Your IdeasGive Claude a specific review job, clear criteria, and evidence to inspect. Turn its objections into useful tests before deciding what to change. · 5 min read
- Zabbix MCP Server: Connect AI to Monitoring InvestigationsAn MCP server for Zabbix lets an AI assistant query alerts, host status, and event history during an investigation. Here is how to set one up, what access to grant, and how to retain sanitized incident lessons for the next responder. · 6 min read
- How to Send ChatGPT a Table Without Losing Its StructurePasting a table into ChatGPT often garbles the columns. Here are three reliable ways to send tabular data and get accurate analysis back. · 5 min read
- How to Share Notion Pages With ClaudeYou share Notion pages with Claude by authorizing access through Claude's connector, selecting the pages to include, and verifying what Claude can actually read. Here is how each step works and what to check. · 5 min read
- AI Teammate: What It Can Do and How to Work TogetherAn AI teammate is an AI assistant you treat as a working member of your project rather than a one-off tool. This guide explains what that means, how to set clear responsibilities, and how to share context so the work continues across sessions and people. · 5 min read
- AI Ad Copy Generators: Draft Variants You Can TestUse an AI ad copy generator to explore clear, distinct messages from approved claims. Review channel constraints and landing pages, then preserve what the test actually taught you. · 5 min read
- AI Agents for Designers: Briefs, Feedback, and DecisionsAI agents help designers analyze briefs, organize client feedback, and document creative decisions. The design judgment stays with the designer. · 5 min read
- Use an AI Writing Assistant to Draft Documents FasterTurn verified notes into a useful document with a clear brief, an approved outline, and a focused human review. Keep the requirements available when revisions move between AI apps. · 5 min read
- AI Consulting and Implementation: From Pilot to Working ProcessTurn an AI engagement into a workflow your team can operate. Define the scope, test the pilot, and agree on a handoff before expanding it. · 5 min read
- AI Document Language Converter: Translate and Check MeaningTranslate a work document with AI, then check terminology, numbers, omissions, and layout before anyone relies on the result. · 5 min read
- AI for Accountants and Auditors: Uses and Review ControlsHow accountants and auditors use AI for preparation, research, and documentation while maintaining human judgment at every decision point, with a record of responsibilities and decisions outside the financial system. · 5 min read
- AI for Biology Research: Literature, Notes, and VerificationBiology researchers can use AI to navigate literature, organize experimental notes, and compare findings across papers. The underlying claims still need checking against the primary sources. Here is a practical workflow and how to retain source links and project context across sessions. · 4 min read
- AI for Odoo Development: Build With Business ContextAn AI assistant can write Odoo modules, views, and business logic faster than you can type them, but it needs clear context about your version, installed modules, and business rules to produce code that fits. · 6 min read
- AI Memory Plugins: What to Check Before Connecting OneChoose an AI memory plugin by what it saves, where that information goes, and whether you can correct it. Use this checklist to test a tool before giving it real work. · 8 min read
- AI Policy for Law Firms: A Practical TemplateAdapt a law firm AI policy for approved tools, confidential information, source verification, supervision, and client communications. · 5 min read
- AI Programs for Beginners: Start With One Useful TaskChoose an AI program by the work you want to finish. Compare approachable starting points, test their answers, and keep the instructions that help. · 8 min read
- AI to Respond to Emails With the Right Thread ContextDraft a reply that reflects the latest message, earlier agreements, and what you can actually promise. Use a small context packet and a human review before sending. · 5 min read
- AI Workflow Nodes: Inputs, Models, Tools, and Review StepsAn AI workflow is a graph of nodes, each with a job and a data contract. Understanding what each node type does, what it passes on, and where things go wrong makes workflows easier to build and debug. · 6 min read