Browse agent skills

TypeScript client for the Gradium Voice AI API (TTS, STT and S2S over WebSocket)

Unified LLM API with automatic model discovery and provider configuration

Host SDK for tg-support: connect a Telegram bot so each user gets a topic in your support forum group and operators reply there.

Dissora - typed, modular Discord bot framework and module manager

n8n node for Microsoft Dynamics 365 / Dynamics CRM on-premises Web API with Windows (NTLM) authentication

The Starting Point for Next-Gen Agents

AI governance gateway and EU AI Act compliance toolkit: links the AI systems you declare with what their traffic shows

Wallaby (Kimi K3) provider for the Vercel AI SDK — OpenAI-compatible, pay-per-token

CLI auditor & generator for /llms.txt, /.well-known/ai-catalog.json (Google Lighthouse 13.5 ARD 1.0), 20+ Schema.org JSON-LD types, 14+ robots.txt AI bot rules, and 25,000-token AI crawler truncation (npx test-ai-ready <url> --write)

Cross-platform Espanso prompt workflow with OpenRouter, Anthropic, Ollama, and LM Studio

Placeholder that reserves the name for TenetRAG, a Python SDK for GraphRAG. Not usable yet.

Evidence-linked project history from local Codex and Claude Code transcripts

Sound-reactive glow for React Native and Expo: a colorful beam along the bottom edge of a chat input or phone screen that rises and blooms with a voice. Skia + Reanimated, runs on the UI thread.

Records each user's AI token usage and cost to your own Supabase project (Cockpitify).

An environment variable parser for nodejs. Unlike dotenv, this parses environment variables instead of loading them.

Speak the Agent Client Protocol to a coding-agent CLI the user has already authenticated. No API key, no third-party adapter.

Holocron plugin for Netlify. Implements the deployment capability against Netlify's REST API, plus exports verifyToken + AUTH_HINT for `holocron auth`.

A domain-retargetable, open-core engine for hybrid retrieval + knowledge-graph RAG: parsing, chunking, embedding, hybrid search, reranking, graph extraction, entity resolution, ontology-driven knowledge, and a capability runtime.

Resumable, file-backed build / review / fix pipelines for Claude Code and Codex, driven by planner, writer and reviewer sub-agents.

GoodMem memory for the Vercel AI SDK: model-facing tools, a retrieval middleware and developer helpers, built on the official @pairsystems/goodmem SDK.

Apple Intelligence helper for any pi model: fm side-chat panel, auto thinking on/off, tiered memory, and on-device QR, speech, and translation tools

Skills is a development skill indexed by Skillful.sh. GitHub stars: 1.

Causal effects on networks with double machine learning: ATE, direct and peer effects with confidence intervals that account for network dependence.

Mattermost is an open core, self-hosted collaboration platform that offers chat, workflow automation, voice calling, screen sharing, and AI integration. This repo is the primary source for core development on the Mattermost platform; it's written in Go an

Web search tool for the pi coding agent — one web_search tool, 10 providers (Tavily, Brave, Serper, Exa, Perplexity, You.com, Parallel, Linkup, SearXNG, DuckDuckGo) with automatic provider fallback. Calls search APIs directly, so a search costs the API pr

Geniffy for the Vercel AI SDK: a memory of each of your users, for any model. Wrap a model, or give it recall and remember tools.

Wuthering Waves echo-screen scanning logic (OCR parsing, layouts, matching, dedupe) shared by Wuthering Tools and Wavescan.

Dictate Pipeup comments with the browser's own speech engine.

Optional native Sherpa speech engine and explicit model preparation for assistant voice.

Duplex voice conversations with application-owned targets, authorization and speech services.