Browse agent skills

A hardened, site-scoped support assistant over an OpenAI-compatible endpoint

OpenRouter Chat Model para n8n com cache de prompt, telemetria de tokenUsage e campo user para rastreio de custo por cliente. Fork de n8n-nodes-openrouter-cache-chat-model.

Arabic-First PDF Extraction, RTL Normalization & Table Chunking Engine for RAG Systems.

AIXX CLI - 让你的AI调用万物。安装：npx aixxai install

Freebuff LLM router - a local OpenAI-compatible /v1 server that proxies free-mode Freebuff models (DeepSeek V4 Flash, MiMo, etc.) for any agent client.

Automatic tool-call review for pi — a local encoder approves safe tool calls and blocks dangerous ones before they run

Metadata telemetry for AI API calls

Open-source WCAG accessibility scanner CLI and GitHub Action (headless Chromium + axe-core). Scans URLs on your own machine's network, prioritized report by impact, optional LLM fix hints.

Playwright testing harness with page-object helpers for @awdlab/jig components.

Retrieve public Sleeper fantasy football data with n8n.

A Claude Code skill that catches commonly-forgotten UX/product details before you ship a screen or feature.

BEAM — distributed graph database syncing over WebSocket, WebRTC, and multicast. Successor to rod.

BrainCell SDK — shared core library with cells, services, and data models for the ITL memory platform

Query and search your markdown notes with context-aware progressive disclosure: SQL over frontmatter, links, and text, plus semantic search and link-graph ranking. No server, no build step

AI development context management system - install and manage skills, rules, context managers and external tool integrations across AI coding tools.

🧠 Persistent memory + 🔍 session search + 🛡️ secret scanning for GSD Pi. Token-aware policy-only memory by default, SQLite FTS5 search, auto-consolidation, procedural skills. 732 tests. Ported from Hermes agent. Fork of chandra447/pi-hermes-memory

Deterministic long-horizon control with Rust/WASM halt and command guards, real tool execution receipts, full checkpoints, and flush-before-compaction.

Passive-awareness swarm bus for agent harness pods — threads, non-blocking mentions, and background watchers that fold teammate discoveries in at step boundaries (AgentRadio, arXiv:2607.28430). Freeze the model, evolve the comms policy.

A Python library for parsing Fountain markup, the screenwriting format

Analyze local AI coding-agent transcripts by minimum execution runtime

Sign, decode and verify JSON Web Tokens, with a speciality in Salesforce's OAuth 2.0 JWT bearer flow and Telephony Integration REST API

Standardized appData bridge for Pinecone webapps

Real-time AI document scanner for React Native — VisionCamera + ONNX Runtime document detection, OpenCV quality analysis, and PaddleOCR (RapidOCR/ONNX) field extraction for passports, driving licences, and ID cards.

OpenCode plugin that classifies bash commands with an LLM before execution: safe commands are allowed, dangerous ones are rejected.

Migrate AI conversations between tools — importers and exporters for @kontourai/thread

Capability-scoped Codex, Claude Code, and Agy adapters for Dharma Agent Fabric

Server-side extension that exposes leitwerk processes through a Telegram bot.

The Amalgm boot shell — the runtime assembled from published @amalgm SDK bricks. Composition and parity ground for the lego-driven engine rewrite.

A tiny metadata-only UsageTap observer for AI usage and cost trials.

Local-first, open-source AI development workspace for Claude Code, Codex CLI, and Gemini CLI