Vibe GuideUNDERSTAND IT. USE IT.
Technical terms, in everyday language
Start with one sentence, then an everyday example and a diagram. Look up what is needed; no memorization required.
755 entries · 27 categoriesNew here? Start with these terms
Browse by category
755 entries; open a category
No match. Try a shorter word, an English name, or clear the filters.
01Project & product22 entries
When describing an idea and defining version one.
02Documents & planning21 entries
When handing an idea to AI and agreeing on tasks.
03AI & models25 entries
When selecting models and evaluating answers.
04Prompts & context25 entries
When instructing AI, attaching material, or changing sessions.
05Agents, skills & MCP28 entries
When extending or authorizing AI task execution.
06AI files & protocols18 entries
07Accounts, costs & limits20 entries
08Knowledge & retrieval24 entries
09Tools & environment30 entries
When creating folders, installing tools, and running commands.
10Code fundamentals36 entries
When reading code explanations or diagnosing errors.
11Architecture & stack31 entries
When choosing the parts needed to run a project.
12Languages & frameworks27 entries
13Project platforms26 entries
14UI & experience32 entries
When describing and reviewing interface design.
15Components & states28 entries
When interacting with controls, loading, or handling errors.
16Browser behavior29 entries
When navigating, storing data, or handling browser errors.
17APIs & networking32 entries
When connecting external services such as login, AI, or payment.
18Data & storage36 entries
When saving, querying, sharing, and backing up data.
19Backend systems28 entries
When organizing services, background work, and concurrent use.
20Identity & security33 entries
When signing in, sharing data, and configuring secrets.
21Git & collaboration32 entries
When versioning, synchronizing, merging, or restoring code.
22Testing & debugging33 entries
When checking AI work, reporting issues, and retesting fixes.
23Build & release29 entries
When moving from local work to real delivery.
24Cloud & containers28 entries
When choosing hosting, configuring servers, or using containers.
25Operations & performance33 entries
When diagnosing slowness, errors, or recovery needs after launch.
26Integrations & business26 entries
When integrating files, email, payment, or location.
27Licenses & ownership23 entries
When reusing code, delivering source, or migrating services.
Similar words, different meanings
Model, tool, agent, and harness
Models generate, tools act, agents pursue goals, and harnesses coordinate runtime and control. These are roles, not interchangeable products.
Prompt, skill, MCP, and plugin
A task instruction, reusable procedure, connection protocol, and extension package serve different purposes. Installation is not unlimited permission.
Model token, access token, and API key
Model tokens measure content; the others are credentials. Do not paste secrets into ordinary task context.
Subscription and API billing
They may be billed separately. Check API balance and limits even with a chat subscription.
Context, memory, knowledge base, and RAG
Current context, persisted memory, a collection, and retrieval-generation differ. Stored information must be retrieved to be used.
Frontend, backend, database, and API
Presentation, business handling, storage, and interfaces have different roles. A success screen alone does not prove persistence.
Website, web app, mini program, and app
Labels can overlap; access, platform capabilities, and distribution differ. Start from real needs.
Library, framework, SDK, and runtime
Reusable code, structural conventions, development kits, and execution support differ; one product may cover several roles.
Git and GitHub
Git tracks versions; GitHub hosts and coordinates repositories. Git does not require GitHub.
Save, commit, push, and deploy
Writing files, recording history, syncing a repository, and deploying are four different actions.
Local preview, preview deployment, and go-live
Location, audience, and purpose differ. Local success does not prove public access.
Authentication and authorization
After identity is established, permissions still need checking. A signed-in user cannot necessarily read others’ records.
Prototype, MVP, and production
A prototype explores design, an MVP tests core value, and production needs operational readiness.
AI claims, test results, and acceptance
Inspect test evidence and real usage separately; an AI summary is not execution evidence.
Build, deployment, and usable behavior
Generating an artifact, starting a process, and completing a user goal are separate checks.
Delete, rollback, and restore
Deletion removes data, rollback changes versions, and restoration recovers saved state; their effects differ.
Cache, database, and backup
Performance copies, business records, and recovery copies serve different purposes. Cache must not be the only essential copy.
RAG, fine-tuning, and long context
RAG retrieves, fine-tuning updates parameters, and larger context holds more material. They solve different problems.
Does this method fit your project, and how can you check?
- When would you use this method?
- When an unfamiliar term blocks a decision.
- How should you choose for your situation?
- Read the plain meaning and decision question, then answer for this project without memorizing the glossary.
- What is easy to misunderstand?
- A definition is not an installation or selection instruction; an unfamiliar technology need not be added.
- How can you check that this works for your project?
- Explain its role in one sentence and return to the action; if unclear ask for a project-specific example.
Try with this project: fill material and prepare the task
Edit the template with actual material, confirm, then copy into the current project conversation. Mark unknowns as undecided and replace examples as needed.
Confirmed · personalized content ready to copy
Which term do you not understand: [fill in]
Original sentence and current step: [paste]
What exactly should this project help you do: [fill in]
Explain how this term affects the current action in plain language with a project-specific example; separate necessary decisions from optional detail.
Explain your conclusions using these points: Plain explanation, project example, when needed/not needed and a concrete next action. Do not install technology merely because its name appeared.Analogies help understanding but do not replace actual rules. Each entry includes context, limits, and further reading.