Claude Code MCP Servers — Wiring Up Real Tools
What MCP is, how to add a server with claude mcp add, three transport types, and the first servers worth installing.
#Claude Code
Articles tagged with #Claude Code
Claude Code Hook Blocks Edit Unexpectedly
A PreToolUse hook keeps rejecting Edit calls with no clear reason, including safe edits. Usually exit-code logic, stdin parsing, or matcher scope at fault.
Claude Code MCP Call Times Out Repeatedly
MCP tool calls (filesystem, GitHub, custom) time out in Claude Code, but a direct curl to the server works. Usually transport, stdio framing, or response shape.
Claude Code Output Truncated by Context Window
A long agent reply gets cut off mid-sentence under context pressure. Usually compaction, max_tokens cap, or runaway tool output eating the budget.
Claude Code settings.json Not Loading After Edit
You edited ~/.claude/settings.json but Claude Code ignores hooks, permissions, or env on restart. Usually JSON syntax, wrong scope, or shadowed precedence.
Claude Code Skill Not Discovered or Auto-Triggered
You dropped a skill under ~/.claude/skills/ but /<name> says not found, or it never auto-fires. Usually a SKILL.md frontmatter, path, or trigger-description issue.
Claude Code Subagent Result Not Relayed to Main Session
Your Task/Agent subagent completed cleanly, but the main session shows nothing or a wrong summary. Usually a final-message format, context-window, or thread-routing issue.
Claude Code Bash Sandbox Blocks an Expected Command
Claude Code refuses to run a command you whitelisted — the Bash sandbox rejects it as unsafe or unapproved. Diagnose permission scope, pattern match, and settings.json wiring.
Claude Code Project CLAUDE.md Not Loading Into Context
Claude Code starts a session but ignores your project CLAUDE.md — conventions, commands, and instructions never reach the model. Diagnose path, scope, and merge precedence.
Claude Code Session Resume Loses Memory of Prior Work
You resume a Claude Code session and it acts like a fresh start — forgetting the plan, the decisions, the half-finished refactor. Diagnose snapshot scope, compaction, and memory layers.
Claude Code Statusline Custom Script Errors or Hangs
Your Claude Code statusline shows an error, stays blank, or freezes the prompt — the custom script is failing silently. Diagnose exit codes, output format, and timeout limits.
Claude Code Permissions Prompt Loop
Every Bash, Edit, Read asks for permission again even after you allowed the same tool — settings.json scope wrong, allowlist pattern too narrow, or hook is re-asking.
Claude Code Token Budget Too Large for One Task
Single task burns 200K+ tokens, hits the context ceiling mid-execution, and forces a summarize. Decompose the task, use sub-agents, or scope reads tighter.
Claude Code Tool Execution Hangs Without Timeout
Bash command runs forever, the agent waits forever — no timeout set means a hung process locks the whole session. Set timeouts, detect hangs, kill cleanly.
Every Claude Code Keyboard Shortcut, Explained
A complete reference for every Claude Code keyboard shortcut: session control, permission modes, input editing, history, IDE integration, and terminal compatibility.
The Complete Guide to Claude Code Skills: Built-in, Custom, Plugin, Team-Shared
A complete walkthrough of Claude Code skills: what they are, how to invoke them, how to write a SKILL.md, when to use hooks or subagents instead, and how to share with a team.
Every Claude Code Slash Command, Explained: /usage to /agents
A complete reference to every built-in Claude Code slash command: session lifecycle, usage and cost, configuration, subagent / hook / MCP extensions, diagnostics, and IDE / git integrations.
Claude Code Accidentally Committed a Secret
`.env` or API key swept into a commit by `git add .`. Rotate first (always), then prevent: ban broad `git add`, install secret scanner, gitignore.
Claude Code Creates a Pile of Unused Helpers
Finishes the task, also adds 5 "might be useful later" utility functions nobody calls. Bound scope in the prompt and in CLAUDE.md, delete on review.
Claude Code Skips or Weakens Failing Tests
Agent makes tests green by adding `.skip`, deleting assertions, or broadening matchers. Forbid test edits, scan diff for skip markers, treat test changes as a separate PR.
Multi-Agent Handoff Prompts for Claude Code Subagents
12 prompt templates to hand work between Claude Code subagents — research → plan → implement → review → ship — without losing context.
Architecture Review Prompts for Layer and Dependency Audits
Architecture review prompts that surface real layering bugs, dependency cycles, and boundary leaks — not generic "consider DDD" advice.
Full Repository Audit Prompts: 15 Templates for Whole-Project Review
Whole-repo audit prompts for Claude Code / Codex — architecture smells, dead code, security risks, dependency drift, test coverage gaps, in one structured pass.
AI Agent Code Review Workflow
Use Claude Code, Codex, or Cursor agents to pre-review your own PRs before sending them to humans.
Claude Code Beginner Guide: Setup, First Refactor, and Mistakes to Avoid
What Claude Code is, how to install it, and how to ship your first refactor — the setup steps, prompt patterns, and common beginner mistakes.
Claude Code — Introduction
Claude Code is a terminal agent — here is what it does and why developers love it.
Project Prompts for Claude Code
CLAUDE.md isn't just a readme — it's your agent's memory. Here's what to put in it.
Claude Code Project Setup — From Empty Repo to First Useful Run
How to point Claude Code at a real project and get it productive in 30 minutes.
Run a Full SEO Audit Using Claude Code
A reusable Claude Code prompt + workflow to run a real SEO audit on your content site.
Claude Code Workflow — Repeatable Patterns
Day-to-day Claude Code patterns that survive past week one — task sizing, commit cadence, CLAUDE.md upkeep, and how to keep agent output trustworthy long-term.
Claude GitHub Integration — A Real Workflow
Connect Claude to a GitHub repo and use it as a research tool over your own code.
Agent Rollback Was Incomplete
Code is reverted, but the agent left behind: stale `dist/`, an applied migration, dangling feature flags, mutated state in third-party tools. Roll back by domain, not by `git reset`.
Claude Code Keeps Asking Unnecessary Questions
You wanted execution; agent stopped every 30 seconds with "should I use TypeScript?" Define defaults in CLAUDE.md + an explicit "only ask for destructive ops" rule.
Claude Code Lost Project Context Mid-Task
Agent suddenly forgets earlier answers, re-reads CLAUDE.md, contradicts the plan. Auto-summarization dropped the load-bearing details. Re-anchor or restart.
Claude Code Ignores Your Audit Report
You fed a detailed audit and Claude Code did its own thing. The report is context, not a contract — convert to a numbered checklist, execute one item at a time.
Claude Code Misunderstands Your Project Architecture
Agent suggests "extract this into a microservice" when you're intentionally monolithic. Codify conventions in CLAUDE.md + do/don't examples from real files.
Claude Code Output Does Not Match Your Original Prompt
The diff is polished code — for the wrong feature. Force a restate-before-execute step; misalignment almost always happens at the plan stage, not the code stage.
Claude Code Plan Mode Diverges From the Plan
You approved a 5-step plan; Claude did 7 steps, skipped one, and refactored adjacent code. Bind execution to the approved plan + post-step status updates.
Claude Code Overwrites Existing Changes
Uncommitted edits gone after agent ran. Local-history recovery, then enforce "commit before agent" + dirty-state checks.
How to Help Claude Code Understand Your Project: 7 Methods That Actually Work
Claude Code editing the wrong files, duplicating existing functions, missing key files? It's almost never the model — it's missing project context. Seven ways to make Claude Code actually read your codebase.
Claude Code Refactor Scope Becomes Too Broad
You asked for a small change, got a 5-file diff. Set a hard editable-file list in the prompt + use `FOLLOWUPS.md` to capture adjacent issues without fixing them.
Claude Code Stops Mid-Task Due to Token / Usage Limits
Halfway through a refactor, Claude Code says you hit a limit. Identify which limit (5-hour / weekly / per-model), save state, pick the right resumption path.
Claude Code Stuck After Partial Execution
Agent did 60% then went silent — no error, no completion. Three flavors: context exhaustion, tool hang, silent abort. Each has its own diagnosis.
Claude Code Stuck in a Loop — How to Break Out
Agent repeats the same edit, oscillates between two states, or chases a flaky test forever. Six loop signatures and the one-prompt fix for each.
Claude Code Edited the Wrong File
Modified a file you didn't intend — usually similar names, or it followed an import chain. Git-revert, then re-scope with allow/deny lists in the prompt.
Multi-Agent Conflict: Two Agents Edited the Same Files
Claude Code + Codex (or two parallel sessions) edit the same function differently and produce a conflict. Isolate scopes with worktrees; never let an agent resolve.
Claude Code Execution Prompts: Plan, Build, Verify
12 prompts to brief Claude Code (or Codex) on real engineering tasks — scoped features, surgical bug fixes, migrations, refactors, TDD, perf work, debugging, code review.
Claude Code vs Cursor: Which Task Belongs to Which
Claude Code is a terminal agent; Cursor is an in-editor copilot. Task fit, debug loop, and rollback cost differ. A decision sheet.
Using Claude Code to Build a Content Site End-to-End
A realistic 2026 walkthrough of building an Astro content site with Claude Code — from repo setup to deploy — including what to delegate, what to keep, and where the model bites you.
Designing Prompts for Website Building (So AI Doesn't Drift)
Keep AI agents on-spec for multi-file website work with this CLAUDE.md template, 3-part prompt structure, and per-task constraint pattern.
What Is Claude Code? Anthropic's Official Terminal AI Explained
Claude Code is Anthropic's command-line AI tool. This guide explains what it does, how it differs from ChatGPT and Cursor, and how to get productive in the first ten minutes.