Reusable AI skills and capabilities for agent workflows
emran
High-performance streaming Markdown text splitter for LLM pipelines and RAG systems. Zero sequence loss for code blocks, tables, links, and videos. 5 built-in strategies + custom. Zero dependencies.
...morerane
A command-line interface for bulk-processing Anki flashcards with LLMs
spanora
OTEL-native observability SDK for AI executions
GitHub Actions
Pa11y CI is a CI-centric accessibility test runner, built using Pa11y
oulianov
Phospho, the LLM analytics platform
GitHub Actions
OpenCode tmux integration: automatically opens subagent panes and renders real-time agent execution with smart layouts
luunix
MCP runtime + package manager + memory layer - One MCP to rule them all
yleflour_pairprog
Immutable, chainable, and type-safe wrapper of Vercel's AI SDK
boqiang.liang
MCP build: init → build → run. Turn your project (@tool, SKILL, n8n) into a standalone MCP npm server (no framework embedding, no package conflict).
...moredhnillhaq
AI-powered CLI tool to auto-generate technical documentation from legacy/undocumented codebases
whamp
Versioned memory extension for the pi coding agent
darrenedwards111
Deterministic action verification for LLM agent security
grimmerk
Transform TypeScript classes into JSON Schema definitions with automatic support for OpenAI, Anthropic, and Google Gemini function calling (tool) formats
...moreanindya007
Manifest-based LLM executor for Studio agents with CLI
eyepaq
Auto-generate /llm.txt and /llm-full.txt for any website
gaurav-nelson
Validate and fix Reviewdog Diagnostic Format (RD Format) - A comprehensive library and CLI tool for validating JSON data against the Reviewdog Diagnostic Format specification
...moreneelratn
Next.js Exhaustive Testing Agent
GitHub Actions
Server-Sent Events (SSE) support for Fetcher HTTP client with native LLM streaming API support. Enables real-time data streaming and token-by-token LLM response handling.
...morekasin-it
Content negotiation for Next.js - serve Markdown to LLMs, HTML to browsers, from a single URL
jorgt
Durable TypeScript runtime for agents and automations. Effects store their results, retries pick up where you left off.
cxy13h
a zero-intrusion LLM call logging library
hangingahaw
Structured typography rules for LLM-based text correction tools
imdigitalashish
Memory-augmented coding agent with persistent learning across sessions. Built on pi.
lacerbi
Electron-specific library for managing local AI model servers and resources