▤
Claude Code Beginner Tips & Tricks
The one-page desktop reference every new Claude Code user wants.
Twelve uniform tiles covering sessions, the closed-loop workflow, file handling, plugins, mid-flight steering, permission modes, agents, memory, and context health — each command and tip verified against the official docs.
✳Claude Code○Free▰VibeVersityBeginner
getting-startedworkflowcommandsmemory
When to use & gotchas
When to use Keep it open while you learn the surface area — sessions, permission modes, memory, and the closed-loop workflow — until the muscle memory sticks.
Gotcha The single biggest beginner win is permission modes: start in plan mode (Shift+Tab) so the agent proposes before it edits, then approve.
▤
Claude Code: What's New
The latest features, from models to autonomy to enterprise.
A six-card map of the most recent Claude Code release wave — 1M context, fast mode, voice, agent view, dynamic workflows, computer use, native install — with version pills showing exactly when each feature landed.
✳Claude Code○Free▰VibeVersityIntermediate
changelogfeaturesmodelsautonomy
When to use & gotchas
When to use Skim it after any gap away from Claude Code — version pills show exactly which release each feature landed in, so you know what's safe to rely on.
◆
Anthropic Agent Skills
Reusable, model-invoked skill packages for Claude.
Anthropic's open collection of Agent Skills — folders of instructions, scripts, and resources Claude loads on demand (document creation, data work, and more). The reference pattern for authoring your own.
✳Claude Code○Free✓OfficialIntermediate
skillsautomationauthoring
When to use & gotchas
When to use Use a skill when you keep re-explaining the same procedure — wrap the instructions and helper scripts once, and Claude loads them on demand.
Gotcha Skills are model-invoked: a vague description means Claude won't reach for the skill at the right moment. Write the trigger conditions explicitly.
◆
Supabase Agent Skills
Database, migration, and security guidance for agents.
Supabase's installable agent skills add critical local-development and security guidance for working against a Supabase project — installed with a single `npx skills add` command.
✳Claude Code○Free✓OfficialIntermediate
skillsdatabasesecurity
npx skills add supabase/agent-skillsWhen to use & gotchas
When to use Install it before letting an agent near a Supabase project — it front-loads the RLS and migration guidance that prevents the expensive mistakes.
⬡
Model Context Protocol — Reference Servers
Filesystem, Git, Fetch, Memory, and more, maintained by Anthropic.
The canonical reference implementations of MCP. Start here to understand the protocol and to wire up filesystem, git, and fetch access for any MCP-capable client.
∗Any tool○Free✓OfficialBeginner
mcpfilesystemgitreference
When to use & gotchas
When to use Start here before any third-party server — the filesystem and fetch references teach you how MCP wiring, scopes, and approvals actually behave.
Gotcha Give a server the narrowest filesystem scope that works. A broad root means the agent can read more of your machine than you intended.
⬡
GitHub MCP Server
Issues, PRs, Actions, and code search from your agent.
GitHub's official MCP server lets an agent read and act on repositories — open issues, review PRs, inspect Actions logs — without leaving the editor. Remote and self-hosted options.
∗Any tool○Free✓OfficialIntermediate
mcpgithubcicode-review
When to use & gotchas
When to use Connect it when your loop involves GitHub — letting the agent read failing Actions logs and PR review comments turns a copy-paste chore into one step.
Gotcha Scope the token tightly. A broadly-scoped PAT hands the agent write access to far more than the one repo you're working in.
⬡
Playwright MCP Server
Drive a real browser — navigate, click, assert — from chat.
Microsoft's Playwright MCP gives agents structured, accessibility-tree-based browser control. Ideal for end-to-end test authoring and verifying deployed pages without brittle pixel clicking.
∗Any tool○Free✓OfficialIntermediate
mcpbrowsertestinge2e
When to use & gotchas
When to use Reach for it when you want the agent to verify a deployed page for real — it reads the accessibility tree, so its checks are more robust than pixel clicks.
Gotcha It needs a publicly reachable URL. A sandboxed agent usually can't hit a local dev server, so point it at a deployed preview instead.
⬡
Supabase MCP Server
Query, migrate, and inspect your Postgres from the agent.
Connect an agent to a Supabase project to list tables, apply migrations, run SQL, and read advisors — a fast loop for schema work when paired with disciplined review.
∗Any tool○Free✓OfficialAdvanced
mcpdatabasepostgresmigrations
When to use & gotchas
When to use Use it for inspect-heavy work — listing tables, reading advisors, drafting migrations — where seeing the real schema beats guessing from code.
Gotcha `apply_migration` writes straight to the target project. Always run schema changes against a staging project first and gate every apply behind your own approval.
⬡
Vercel MCP Server
Deployments, build logs, and project state from the agent.
Vercel's MCP server lets an agent list deployments, read build and runtime logs, and inspect project configuration — closing the loop between a push and knowing whether it shipped.
∗Any tool◐Freemium✓OfficialIntermediate
mcpdeploymentlogsci
When to use & gotchas
When to use Connect it when 'did it actually deploy?' is part of your loop — the agent can pull the latest deployment state and build logs instead of you tab-switching.
⬡
Sentry MCP Server
Pull issues and traces into the loop while you fix them.
Sentry's MCP server surfaces error issues, stack traces, and release context to the agent so a fix can be written and verified against the real production signal.
∗Any tool◐Freemium✓OfficialIntermediate
mcpobservabilityerrorsdebugging
When to use & gotchas
When to use Pull a Sentry issue into the session when you're chasing a production bug — the agent fixes against the real stack trace instead of a paraphrase of it.
❑
Claude Code
Anthropic's agentic coding tool — terminal, IDE, and web.
The agent this whole hub is built around: a coding agent that reads your repo, runs commands, edits files, and ships. Included with Claude paid plans; also available via API.
∗Any tool◐Freemium✓OfficialBeginner
agentcliide
When to use & gotchas
When to use Reach for it when you want an agent that owns a whole task end-to-end — read the repo, run the tests, fix, and open the PR — not just autocomplete.
Gotcha It's only as good as the loop you give it: a clear success check (a test or build it can run) is what lets it self-correct instead of guessing.
❑
Cursor
The AI-native code editor (VS Code fork).
A full editor built around AI: multi-file edits, codebase chat, and agent mode. Free tier with usage limits; paid plans unlock higher limits and frontier models.
▸Cursor◐Freemium✓OfficialBeginner
editoragentautocomplete
When to use & gotchas
When to use Pick it if you want AI woven into the editor itself — inline edits, codebase chat, and agent mode — rather than a separate terminal tool.
❑
GitHub Copilot
In-editor completions, chat, and agents across IDEs.
GitHub's assistant for VS Code, JetBrains, and more — completions, chat, and a coding agent. Free tier available; Pro and enterprise plans add capacity and model choice.
∗Any tool◐Freemium✓OfficialBeginner
autocompletechatagent
When to use & gotchas
When to use A safe default if your team spans VS Code and JetBrains and you mainly want fast, in-line completions with chat on the side.
❑
Cline
Open-source autonomous coding agent for VS Code.
A transparent, open-source agent extension: it plans, edits, and runs commands with your approval at each step, and works with the model provider of your choice.
∗Any tool○Free{ }Open sourceIntermediate
agentvscodeopen-source
When to use & gotchas
When to use Choose it when you want to watch every step and bring your own model/key — its approve-each-action flow makes the agent's reasoning visible.
❑
Aider
AI pair programming in your terminal, git-native.
A command-line agent that edits your repo and makes a clean git commit per change. Model-agnostic and fully open source — a favorite for scriptable, reviewable workflows.
∗Any tool○Free{ }Open sourceIntermediate
cligitopen-source
When to use & gotchas
When to use Great when you value a clean git history — it makes one focused commit per change, so every AI edit is easy to review and revert.
❑
Continue
Open-source, configurable assistant for VS Code & JetBrains.
An open-source autopilot you fully configure — bring your own models, rules, and context providers. Strong choice when you want control over what the assistant sees.
∗Any tool○Free{ }Open sourceIntermediate
autocompletechatopen-source
When to use & gotchas
When to use Reach for it when you need to govern exactly which models and context an assistant uses — useful for privacy-sensitive or self-hosted setups.