CODINGVILA COLLECTION

AI utilities

Estimate tokens, inspect prompts, prepare datasets, and generate model-neutral AI formats.

15 focused tools

EXPLORE BY PURPOSE

Choose a ai utilities workflow

Open a complete collection, or jump directly to one of its recommended tools.

4 toolsToken and prompt analysis

Estimate tokens and inspect prompt structure, length, repetition, and risks.

5 toolsPrompt engineering

Build portable prompt formats, examples, schemas, and SDK snippets.

3 toolsAI datasets and conversations

Validate, convert, deduplicate, and redact training and conversation data.

3 toolsAI planning calculators

Estimate context fit, chunking, vector storage, and formula-based AI usage.

ALL UTILITIES

Browse every ai utilities tool

Open a utility below, or search the complete catalog by task, action, or format.

AI Token Counter (Approximation)estimate tokens from Unicode text using transparent character and word heuristics.

Prompt Length Analyzermeasure prompt characters, words, lines, approximate tokens, and context usage.

Prompt Injection Pattern Scannerflag common instruction-override, secret-extraction, and role-confusion phrases using local heuristics.

Prompt Diff and Version Comparercompare two prompt versions line by line without sending their contents anywhere.

Prompt Format Converterconvert role-based message JSON into ChatML, Llama 3, Alpaca, Vicuna, or Mistral text templates.

Few-Shot Example Formatterformat JSON input/output examples into consistent labelled prompt demonstrations.

System vs User Message Splittersplit a labelled prompt into structured system and user message JSON.

Prompt to SDK Code Generatorgenerate syntactically valid Fetch, Python requests, or cURL templates with explicit placeholders.

Structured Output Schema Generatorgenerate JSON Schema, Pydantic, or Zod structure from a representative JSON object.

Fine-Tune JSONL Dataset Validatorvalidate generic chat-message JSONL structure, roles, content, and duplicate records.

Conversation JSON to Markdownconvert a role/content conversation JSON array into readable Markdown.

Prompt / Response Pair Deduplicatorremove exact duplicate prompt-response JSONL records while preserving first occurrence order.

Context Window Fit Calculatorestimate whether prompt, response reserve, and overhead fit within a chosen context window.

RAG Chunk Size and Overlap Calculatorestimate chunk count and duplicated tokens from document length, chunk size, and overlap.

Vector Index Storage Calculatorestimate raw embedding-vector storage from vector count, dimensions, and numeric precision.

Private ai utilities for everyday work

Estimate tokens, inspect prompts, prepare datasets, and generate model-neutral AI formats. Each page explains supported input and produces a real result without a server queue.

Why browser-based processing matters

Local processing reduces delay and avoids sending working data to a conversion service. Your browser and device still set practical memory limits, so review large results before production use.