Skip to content
AI Dev Toolkit.
Esc
  • AI Token CounterCount tokens for GPT, Claude, Gemini, DeepSeek, Qwen and more.Tool
  • LLM API Cost CalculatorEstimate per-request, daily and monthly API costs.Tool
  • AI Model ComparisonCompare prices, context windows and features across models.Tool
  • AI Model Pricing PagesSpecs, real costs and cheaper alternatives for popular models.Tool
  • Context Window CheckerSee whether your text fits each model's context window.Tool
  • Subscription vs API CalculatorFind out whether a chat plan or the API is cheaper for you.Tool
  • GPU / VRAM CalculatorCheck how much VRAM a local model needs and which GPUs fit.Tool
  • Claude Code Error DatabaseExact Claude Code error messages with tested fixes.Tool

Category

Tools for Claude Code, MCP and coding agents

Set up coding agents properly and fix them when they break: project memory files, MCP server configs for every editor, and a database of Claude Code errors with tested fixes.

Start here

Pick a tool by what you need

Tools

Every tool in this category

Coming next: Claude Skills Directory, MCP Server Directory, Claude Code Hooks Builder, AI Rules Converter.

About

How these tools fit together

Coding agents work best with two kinds of setup. A project memory file (CLAUDE.md for Claude Code, AGENTS.md for most other agents) tells the agent your commands, conventions and the traps in your codebase. The CLAUDE.md generator builds one from a short form and keeps it brief, because the file is sent with every request.

MCP servers give an agent tools: your issue tracker, a browser, a database. Every app stores MCP settings differently: VS Code uses a servers key with inputs for secrets, Cursor and Devin Desktop (formerly Windsurf) read secrets with ${env:NAME}, and Claude Desktop only takes local servers in its config file (remote ones are added as connectors). The MCP config generator writes the right format for each, with secrets kept in environment variables.

When something goes wrong, the error database has the exact messages Claude Code prints, what they mean and fixes for each operating system, checked against Anthropic’s documentation and the Claude Code build itself.

Read

Guides

FAQ

Questions people ask

Do these tools work with agents other than Claude Code?

Yes. The memory-file generator writes AGENTS.md, which Codex, Cursor, Copilot and others read, and can point CLAUDE.md at it so both stay in sync. The MCP config generator covers Claude Code, Claude Desktop, Cursor, VS Code and Devin Desktop.

Are my API tokens safe in the MCP config generator?

The generator runs in your browser and never sends what you type anywhere. It writes configs that read tokens from environment variables or the editor’s secret prompts, so you don’t paste secrets into files you might commit.

How are the Claude Code error fixes checked?

Each message is matched against Anthropic’s error reference and the text inside the current Claude Code build, and each fix follows Anthropic’s documentation. Steps that were run on a real machine say so, and every page shows the version and date it was checked.

Is Claude Code cheaper on a plan or on the API?

It depends on how much you use it. Light use is often cheaper on the API; daily heavy use usually favours a plan. The subscription vs API calculator works it out from your own usage.