This implementation demonstrates using /chat/completions as a tool uses progress notifications that show the intermediate generated tokencount and a message until the full chat completions answer has arrived.
Sentencepiece tokenization for natural language processing, JS version.
Count the number of OpenAI tokens in a string. Supports all OpenAI Text models including GPT-5, GPT-4, GPT-3.5-turbo, and specialized models
This SDK enables seamless interaction with Mint Club's smart contracts, offering both read and write capabilities across multiple blockchain networks.
Context compression for AI agents. Fetch -> Extract -> Convert pipeline without LLM dependency.
AI prompt skill engine for [OpenPencil](https://github.com/ZSeven-W/openpencil) — phase-driven prompt loading with intent matching, token budgets, and design memory.
Weaveai core framework - framework-agnostic AI operations
Smart Handoff core library — handoff schema, builders, compression, adapters
Javascript token counter for GPT-3
A package to make easier to count tokens during data analysis
Prioticket Design Tokens for Material Web Components - CSS custom properties with state layers and comprehensive token coverage
High-level binding to tiktoken-rs for Node.js and browsers
A CLI tool to count LLM tokens in text files.
基于@musistudio/claude-code-router的增强版CLI路由工具 - 支持统一路由引擎、外部规则加载、智能日志系统、自动配置迁移和GLM思考模式
Lightweight trimmed down encoder/decoder/tokenizer/token counter for gpt3 that is compatible with both node and browser environments
Custom skills for Pi coding agent
Hush: A Semantic Security Gateway for AI Agents. Redacts PII from prompts and tool outputs locally before they hit the cloud.
Core GDD framework for graph resolution, validation, and health scoring
Using OpenAI GPT models, from a description of a software system or other entity diagram, generate a design diagram image or PDF.
Deep PR review extension for Pi with deterministic context packing and Responses API streaming
A Unicode-aware token segmentation and counting library
n8n节点插件：基于jieba-wasm的中文分词功能
URL in, clean Markdown + structured metadata out. Built for AI agent tool use.
MCP server for sending messages via Discord webhooks
Parallel CLI that counts GPT-style tokens across your codebase
entelix vendor-accurate TokenCounter wrapping HuggingFace `tokenizers` — Llama / Qwen / Mistral / DeepSeek and any tokenizer.json source
entelix vendor-accurate TokenCounter for OpenAI BPE encodings (cl100k_base, o200k_base, p50k_base, r50k_base) — wraps tiktoken-rs with eager preload
Approximate token counts for LLM messages, system prompts, and tools. Zero-config chars/4 heuristic, BYO tokenizer, content-block aware (text/image/tool_use/tool_result/document). One serde_json dep.
Fast token and cost usage statistics CLI for Claude Code, OpenAI Codex, Cursor, and Grok
Axum middleware for Cloudflare-style Markdown for Agents content negotiation
External service integrations (GitHub, Slack, Jira, Anthropic)
Semantic memory with SQLite and Qdrant for Zeph agent
AI-Protocol execution runtime core (protocol, client, pipeline, transport)
Protocol Runtime for AI-Protocol - A high-performance Rust reference implementation
Rust SDK for the Codex CLI — spawn and stream codex exec sessions
Rust SDK for the Gemini CLI — spawn and stream gemini execution sessions
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.