The fenbs blogAll articles
Page 66 of 99
- How to Use Cursor Rules in GitHub CopilotMove the useful content of your Cursor rules into instruction files Copilot supports. Translate when each rule applies, then check that it loads for the right task. · 5 min read
- Cursor Rules for TDD: Keep the Red, Green, Refactor LoopCursor rules that hold the assistant to the TDD cycle: write a failing test, make it pass with minimal code, then refactor. fenbs stores the task acceptance criteria and the lessons from failed attempts. · 4 min read
- Cursor Rules for Vue: Components, State, and TestsCursor rules for Vue that cover component API conventions, reactivity patterns, state ownership, and test requirements, with fenbs keeping the product decisions and naming conventions that should survive a model change. · 4 min read
- Gemini Gems for Education: Reusable Teaching and Study RoutinesGemini Gems let educators and students define a repeatable assistant with specific instructions, tone and scope. Here is how to create useful Gems for teaching and studying, what to store outside them, and how fenbs preserves the prompts, learning notes and next tasks that survive when you switch tools. · 7 min read
- Gemini Memory Across Chats: Carryover and ControlsGemini can reuse information across eligible conversations, but instructions, past-chat memory, and activity controls do different jobs. Here is what to review. · 5 min read
- Turn Off Gemini Memory: What Changes and What StaysStop Gemini from using past chats for personalization, check the separate instructions switch, and understand what you have left in your account. · 5 min read
- GitHub Copilot for Kubernetes: Draft and Review ManifestsGive Copilot explicit deployment requirements, inspect every generated setting, and validate against the intended environment before approving a rollout. · 5 min read
- How to Give Claude Background Information That HelpsGive Claude the background a new colleague would need: the goal, audience, reliable sources, constraints and decisions. Keep that record small enough to check and easy to update. · 5 min read
- How to Stop Claude Being Lazy or Leaving Work UnfinishedClaude leaves work unfinished when the scope is unclear or the task lacks explicit completion criteria. Here is how to define done, set verification steps, and continue from where a session stopped. · 5 min read
- Intellectual Property and Knowledge Management in AI WorkKeep ownership, permissions and source links attached to the knowledge your team reuses. Give assistants clear instructions before a useful reference becomes an unauthorized copy. · 5 min read
- Kiro Agent Memory: Keep Useful Context Across SessionsUse Kiro steering for explicit project guidance, inspect learned memory, and test a fresh session. Keep shared project decisions separate from a private conversation. · 6 min read
- Knowledge Management in Manufacturing: Preserving Improvement WorkManufacturing teams lose process-improvement rationale and project lessons when people move on. Here is how to keep that knowledge working alongside controlled documentation, with and without AI. · 7 min read
- LLM Comparison: Choose a Model for Your Actual WorkBenchmark comparisons tell you how models perform on standardized tests. This guide shows how to compare LLMs on the tasks you actually do — writing, research, analysis and coding — and how to keep your evaluation decisions portable across tools. · 5 min read
- Xero MCP Server: Connect AI to Bookkeeping WorkflowsA Xero MCP server lets an AI assistant query invoices, reconcile transactions, check account balances, and pull financial reports through a standard tool call. Here is how to connect it, set the right access boundaries, and keep bookkeeping process rules and follow-up tasks outside the accounting system. · 4 min read
- ChatGPT Projects in Codex: How to Hand Off the WorkMove the files, instructions, and approved conclusions a coding task needs. Check the working folder and current state before asking Codex to make changes. · 5 min read
- Open-Source MCP Gateways: Compare Routing and Access ControlsAn MCP gateway sits between AI clients and the servers they call, routing requests, enforcing access rules, and centralizing audit logs. Several open-source options exist. This guide explains what to look for, compares the main approaches, and shows where fenbs fits as the shared work record behind the gateway. · 5 min read
- What Is a Shared AI Brain? One Work Record for the Whole TeamA shared AI brain keeps the team's projects, decisions, rules and progress in one record that every connected assistant can read and update. Here is what belongs in it, how to choose a tool and how to start. · 8 min read
- AI Agents for Google Ads: Tasks, Checks, and ApprovalAI can assist with Google Ads research, reporting, and change proposals, but budget decisions and creative approvals belong with a person. Here is how to use AI effectively while keeping spend under your control. · 5 min read
- AI App Creators: From an Idea to a Reviewable PrototypeChoose an AI app creator by the application you need, the changes you can review, and the work you can maintain. Start with a small prototype and explicit acceptance checks. · 8 min read
- AI Candidate Screening: Criteria, Evidence, and Human ReviewUse AI to organize application evidence against approved job criteria. Keep people responsible for screening decisions, accessibility, and checking unequal outcomes. · 5 min read
- AI-Driven Candidate Sourcing: Build a Search People Can ReviewTurn an approved hiring brief into useful search queries, review the evidence behind potential matches, and preserve reusable sourcing instructions. · 5 min read
- AI for Architects and Designers: Practical WorkflowsAI supports briefs, research, option review, and client presentations for architects and designers. Here is how to use it across a project and keep requirements, sources, and approved decisions organized. · 7 min read
- AI for Product Design: Briefs, Concepts, and ReviewProduct designers use AI to turn research and constraints into concept briefs and to sharpen prototype feedback — but the design rationale and approved revisions need to stay with the project. · 5 min read
- AI for Research and Development: From Questions to ExperimentsUse AI to organize evidence, challenge assumptions, and prepare experiments. Keep people responsible for the research decisions and the results you report. · 5 min read
- AI Summary Generators: Choose a Tool That Keeps the MeaningAI summary generators can compress any text into shorter notes. Here is how to compare them by accuracy, length control and source handling, and how to store reusable summary instructions for consistent results. · 7 min read
- The Best Way to Use AI for AssignmentsStart with your course policy, use AI for a specific learning task, and check every contribution. Keep the thinking, evidence, and final submission under your control. · 5 min read
- Can Claude See Images? What It Can Read and What to CheckClaude can analyze photos, screenshots, charts, diagrams, and PDFs that contain images when you share them directly. Here is what it reads well, where it struggles, and how to use image analysis for work. · 6 min read
- ChatGPT Enterprise Agents: Roll Out Shared Work SafelyStart with a narrow job, confirm access, and name the people who own instructions and review. A shared rollout record keeps approved decisions available when the team changes AI apps. · 5 min read
- ChatGPT for Zendesk: Drafting and Reviewing Support RepliesA step-by-step workflow for using ChatGPT to draft Zendesk support replies faster, with approved context, human review before sending and no customer data left in prompts. · 5 min read
- ChatGPT Prompts for Customer Service: Replies and EscalationUse approved policies and verified case facts to draft replies, summarize issues and prepare escalations. Copy these prompts and keep the review rules with the work. · 5 min read
- ChatGPT Prompts for Nonprofits: Programs, Grants and OutreachUse practical prompts to test program plans, organize grant requirements, and draft outreach from verified evidence. Keep the mission clear, protect personal information, and leave commitments to people. · 5 min read
- ChatGPT vs. Dola AI: Conversations or Scheduling Help?ChatGPT handles conversation, writing, analysis, and general questions. Dola AI focuses on calendar management through messaging. They serve different needs and are often better used together. · 4 min read
- Claude Code and Xcode: Setup, Changes, and Build ChecksClaude Code runs as a CLI alongside Xcode. Here is how to set up the workflow, give Claude Code your project context, verify edits through builds and tests, and record what you learn in fenbs. · 5 min read
- Connect Claude to Atlassian: Jira and Confluence OptionsSeveral MCP servers and third-party connectors let you bring Jira and Confluence content into Claude sessions. Here is how to evaluate the options, verify the access you need, and keep shared decisions available across assistants. · 5 min read
- Claude for Unity: Scripts, Debugging and Project ContextUse Claude to write and debug Unity C# scripts by supplying the engine version, project constraints and test results. Here is a repeatable workflow for Unity developers, with fenbs keeping the task brief and lessons between sessions. · 5 min read
- Claude vs ChatGPT for Trading ResearchClaude and ChatGPT can both support trading research by helping interpret filings, test market hypotheses, and check reasoning. Neither provides real-time prices or market data. Here is how to compare them for the research work traders actually do. · 7 min read
- Claude vs Gemini for Excel: Compare Spreadsheet WorkflowsA practical comparison of the routes Claude and Gemini offer for Excel files, formula work, and checking results, with guidance on keeping workbook requirements and verified calculation decisions portable. · 5 min read
- Claude vs Grok: Compare Them on Your Own WorkClaude and Grok take different approaches to everyday tasks. This guide compares them on writing, research, reasoning, and source checking so you can pick the one that fits your work. · 7 min read
- Codex CLI: Go Back to a Previous Message or Task StateCodex CLI does not have a command to replay a previous message. Here is what "going back" actually means in a terminal coding agent, how to undo file changes, and how to preserve checkpoints so you always have a safe state to return to. · 4 min read
- Company Brain With Claude: Give Every Session the Right BriefBuild a useful company brain for Claude with a short team brief, approved rules, source links, and current work. Keep the shared record clear enough for the next person to trust. · 5 min read
- Copilot: View Memories and Review What It RemembersFind the memory controls for your Copilot account, review saved details, and correct suggestions before treating them as project facts. · 5 min read
- Cursor Rules Hierarchy: Scope, Precedence, and ConflictsCursor loads rules at different scopes — global, project, and subdirectory — and resolves conflicts when two rules apply to the same file. Learn how the hierarchy works and how to keep the decisions behind it in one place. · 6 min read
- Does Claude Delete Chats? Saving, Retention, and RecoveryClaude has saved history, user deletion, incognito conversations, and organizational retention controls. Check which applies before assuming a missing conversation can be recovered. · 5 min read
- Enterprise AI: Shared Rules, Ownership, and RolloutEnterprise AI means using AI within the responsibilities and systems of an organization. Choose the work, access, owners, and review process before expanding the rollout. · 5 min read
- External Knowledge Management: Keep Sources You Can TrustBuild a small, useful record of outside references, the claims they support, and when to check them again. Keep that context available when an AI assistant picks up the work. · 5 min read
- Does Gemini Pro Have a Limit? Usage and ContextGemini Pro has usage limits that vary by plan and access path. App subscriptions cap message volume; the API imposes rate limits and context window bounds. Here is how each limit works and what to do when you hit one. · 5 min read
- Google AI SDK: Build a Work Assistant With Shared ContextThe Google AI SDK lets you call Gemini models from Python or JavaScript with a few lines of code. Here is how to set it up, add tools, supply project context, and connect to fenbs over MCP for a shared record across sessions. · 4 min read
- Government Knowledge Management for AI-Assisted TeamsGovernment teams need institutional knowledge to survive staff transitions and policy changes. Here is a workflow for capturing approved guidance, project decisions, and lessons where AI tools are permitted. · 7 min read