All tools
16 tools, each doing one job. Every one processes your input in your browser.
8 tools
AI & LLM
Tools for building and debugging LLM applications.
- Token & Cost CalculatorEstimate token usage and monthly API costs.
- JSON Schema BuilderBuild a schema visually and export it as a tool definition.
- Structured Output ValidatorCheck model output against a schema, with the path that broke.
- Prompt DiffSee which instructions changed, and what the change costs.
- MCP Config ValidatorCatch the mistakes that break an MCP config in practice.
- RAG Chunking PlaygroundSee the chunks your settings actually produce, before you index.
- Embedding Cost EstimatorWork out what embedding a corpus costs before you index it.
- Multi-model Cost ComparisonPrice one workload across several models, side by side.
2 tools
Generators
Produce configuration and instruction files from structured input.
3 tools
JSON
Format, convert, compare and turn JSON into something else.
3 tools
Security
Find credentials and secrets before they reach a commit.
Choose by the job you need to finish
Use the JSON Schema Builder when you are defining the shape of data an application or model must return. Use the Structured Output Validator when you already have a schema and need to learn exactly why a JSON response fails it. They solve opposite sides of the same contract: writing it and checking it.
Use Prompt Diff when two instructions look almost identical but behave differently. It highlights word or line changes and estimates the token difference. Use the MCP Config Validator when the problem is a server configuration instead: it checks the fields and values that commonly stop an MCP client from connecting.
Use the RAG Chunking Playground before committing to chunk size and overlap. It shows the chunks in the browser so you can see split sentences, repeated context and fragments that are too small to retrieve usefully. Use the Instruction File Generator when the output you need is a practical AGENTS.md or CLAUDE.md draft for a repository.
Use the Secret & API Key Scanner before sharing a diff, configuration file or log excerpt. It looks for credential-shaped text and masks matches. A clean scan is not proof that a file contains no secret, but it is a useful check before a commit or paste leaves your machine.
What every available tool shares
Each tool does its processing locally in the browser and is usable without an account. The interface keeps input before output, preserves pasted text when validation fails and explains where a problem occurred and what to change. Choose the smallest tool that matches the task; there is no setup step or shared project state to maintain.