Search

Rewrite AI-generated text to sound human. 3-pass editing system with 36+ banned words, 10 structural patterns, and a quality checklist. Works with any LLM — Claude, ChatGPT, Gemini, Cursor, Windsurf. Also available as a Claude Code plugin.

Tokenflood is a load testing framework for simulating arbitary loads on instruction-tuned LLMs.

LLM visibility monitoring for brands - track how your brand appears in AI-generated responses

Storage connector for reading Spark event logs from HDFS, S3, and GCS

ÇEDD Çözüm pediatric clinical calculators - CLI interface

OpenClaw plugin that logs LLM inputs and outputs to a local JSONL file.

git-spec - Reverse engineer GitHub repositories into comprehensive documentation

Security scanner for ClawHub / OpenClaw AI skills — static analysis + LLM-based evaluation

An Ant Design Claude skill package

Type-safe prompt management with automatic optimization for LLMs

TypeScript wrapper around @anthropic-ai/claude-agent-sdk with progressive disclosure API

Adds a web_search tool powered by Anthropic's native web search — no third-party API key needed. Always runs via claude-haiku-4-5 for fast, cheap retrieval with source URLs.

Pi extension for loading skills on demand from any location

Deterministic LLM workflows you can replay — explicit states, strict schemas, local logs.

Enhanced ManifestStaticFilesStorage for Django

AI-powered workflow automation CLI — build, run, and automate workflows from the terminal

On-device OCR for React Native. Offline text recognition using Apple Vision on iOS and Google ML Kit on Android, built on Nitro Modules.

RAG in 3 lines of code — load documents, ask questions, get answers

Wovv AI Assistant Cross-platform Package

Model-agnostic LLM evaluation that feels like pytest.