npm agents

> A minimal agent framework with origin-centric computation and modular equipment system

Framework-agnostic vault-entry-as-tool resolver for deny.sh. The shared security core behind every deny.sh agent-framework adapter.

MCP Server exposing AgentTrust tools — lets any Agent framework query trust scores via npx

TypeScript SDK for the Chidori agent framework — HTTP client plus agent authoring types.

Minimal Claude Code agent framework for building custom pipelines

Scaffold agentic development documentation & configuration into any project

Agentic Lab Skills

One guided CLI to install, update, and manage Agentic OS + Skill Systems.

AI project initialization kit for agentic coding workflows.

ACE — Agentic Collaborative Engineering: Claude Code agent team installer

CLI to manage Agentic Workflows - install skills, workflows, and processes for AI agents

Visualize and follow your agentic-surfaces workflows in the browser — run `npx agentic-surfaces` from your repo to open the live dashboard.

Curated QE skill pack — 20 quality engineering skills from agentic-qe for Claude Code agents.

Broadcast your agentic coding session to the world in real-time

Agentic Time: measure an AI agent's internal time and 7-state health from per-channel state deltas, in the browser, on the edge, or in Node — via a tiny WASM build of the emergent-time Rust crate.

OpenClaw plugin for Radware Agentic Protection

Repository-native agent memory, skills, AGENTS.md, llms.txt, and knowledge workflows for coding agents.

Tenant isolation and per-tenant usage metering for LangGraph.js checkpointers and stores

Frontal Agents SDK — define, deploy, and observe AI agents with LangChain, LangGraph, and Vercel AI SDK integration

An autonomous, AI-powered CLI tool for automatic build error detection and repair.

A portable spec and live Kanban board for orchestrating AI agent workflows.

Self-improving AI agent CLI — opencode TUI + skills. Installs the arcana binary.

MetaHarness — mint a custom AI agent harness from any repo. Browser Studio + `npx metaharness` CLI. Runs on Claude Code, Codex, pi.dev, Hermes, OpenClaw, RVM.

Runtime authorization guard for AI agent tool calls.

Core library for Signet - portable AI agent identity

One-command setup package for Base MCP AI Agent infrastructure

Git for AI agent memory — TypeScript SDK for MemForks on Sui

Replayable, recoverable, remotely controllable AI Agent workspace.

Axiom Lattice Web is the frontend web application for the Axiom Lattice AI agent platform.

Runner daemon for Yggdrasil — registration, heartbeat, task execution, Realm lifecycle relay, and LLM/agent capability presets

Typed in-process tool definition, registration, and permission-scoped execution for LLM agent loops

Shared LLM agent engine + Vue/Quasar UI for Tauri apps (chat, journal, trust-tier approval).

Tripwire security for LLM agent pipelines — honeytokens that fire when an agent acts on an injected instruction. v0.1 coming soon.

Local-first CLI for installing host-specific LLM wiki skills.

Headless agent runtime for the HoloMesh — multi-LLM-provider, brain-mounted, cost-guarded. Closes W.087 vertex B.

Static-analysis MRI for codebases: a Tree-Sitter dossier engine and deterministic PreToolUse safety gate for coding agents. Language-agnostic core, optimized for TypeScript/JavaScript.

OpenViking memory adapters for LangChain.js and LangGraph.js — retriever, store, agent tools, chat history, and context middleware.

Nebuly-AI server side SDK for JavaScript

TypeScript SDK for agent portability, discovery and trust based on ERC-8004

LangChain.js integration for Prefactor observability

Deno Sandbox backend for deepagents

Qode — LLM provider abstraction layer (Anthropic, OpenAI, Google, Ollama, etc.)

TypeScript orchestrator that runs coding agents (Claude, Codex) in per-issue Gondolin microVMs over ACP.

Local AI agent daemon — voice control, screen vision, any-provider LLM routing, with a permission gateway on every action

OpenAI-compatible embedding provider for @vivantel/rag-core

Shared MetaMask agent skills installer and sync CLI.

MiMo Code: Where Models and Agents Co-Evolve

Wire an autonomous agent to a Solid pod — NIP-98 authenticated read/write with the agent's own key.

Local-first, agent-native control plane for ComfyUI — MCP server + autonomous sidebar agent that drives your live graph in natural language on ANY LLM: Claude/ChatGPT/Gemini on your subscription (no API key), free local models via Ollama (fully offline),

An autonomous software company: CEO (you) + CTO/CPO/CMO leading 15 specialist agents. Run /cto and the CTO takes over — digests the project, fixes fires, improves the product, reports back with a roadmap.

Single source of truth for Redis channel names, key patterns, and message shapes across the cc-suite (cc-tg, cc-agent, cc-agent-ui)

A deterministic stop condition and code-graph context layer for AI coding agents: gives agents a code graph to make changes, then blocks only net-new detector-backed regressions at the stop boundary, with no model in the gate.

Suveren gateway — local agent gateway built in compliance with the Human Agency Protocol (HAP). Runs the UI, control plane, and MCP server in one Node process.

Lightweight container orchestration for Claude AI agents. Pure JavaScript ESM.

Obsidian plugin for file-backed AI agents, task scheduling, channels (Slack, Telegram, Discord), heartbeat, and interactive chat.

Qode — Autonomous Coding Agent CLI

Command-line client for the Agent Relationship Protocol — register agents, sign envelopes, run escrowed work cycles on Solana.

Universal AI bridge for Obsidian vaults - connect any MCP-compatible assistant

Access AI models, email services, and project thumbnail capture through vly.ai and freebuff.com without managing multiple API keys or accounts

Multi-tiered memory and knowledge base plugin for TUI coding assistants

The SDK for nebula, a Casper-native policy-aware AI treasury agent: brain, local memory + storage, the policy/approval engine, Casper account identity, and the on-chain registry client

Local-first AI coding CLI with embedded model router, multi-agent orchestration, and codebase indexing

Sentinel Security Platform — AI-powered multi-agent security engineering framework

Local-first Goal Loop OS for long-term AI work across terminal, web, desktop, mobile app, and messengers.

Multi-agent orchestration engine for Claude, Codex, and Gemini

Scaffold PR review, security analysis, and multi-agent DAG starters on @open-multi-agent/core.

SuperInstance T-Minus Cue Dispatcher — temporal orchestration for multi-agent AI fleets

Dynamic workflow engine for multi-agent LLM orchestration

MCP server exposing Glorp Garage orchestration — namespaces, workspaces, sessions, and the multi-agent roster — as tools for any MCP-capable agent.

Autonomous multi-agent development orchestrator with TDD, security, and craftsmanship

Lark/Feishu channel for the eve agent framework

Thin agent-framework adapters for Relayfile. One connect() call, ready tools, proven writeback lifecycle.

OuTre is a fully modular AI Trading Agent Framework designed for **Solana-based trading automation**. It enables traders to **create AI agents**, **execute strategies**, and **manage portfolios** via JSON configurations.

Jira Product Discovery (JPD) helpers for [agentic-surfaces](https://github.com/dalebaldwin/agentic-surfaces) workflows.

The command-line runner for [agentic-surfaces](https://github.com/dalebaldwin/agentic-surfaces) — a local-first, agent-in-the-loop workflow engine. Loads a project's `agentic-surfaces.config.yaml` + `workflows/*.yaml` + `agents/*.md`, runs the scheduler,

The agent runner for [agentic-surfaces](https://github.com/dalebaldwin/agentic-surfaces) — a thin wrapper over the Claude Agent SDK used by `agent.run` nodes. Supports capability **profiles** (`constrained` / `surface` / `full`), structured **decision** o

The engine core for [agentic-surfaces](https://github.com/dalebaldwin/agentic-surfaces) — workflow schema + loader, the node executor and handler registry (`task.http`, `task.transform`, `task.branch`, `task.foreach`, `task.cache-unseen`/`task.cache-mark`

The dashboard server for [agentic-surfaces](https://github.com/dalebaldwin/agentic-surfaces) — an HTTP + SSE API (`/api/workflows`, `/api/run`, `/api/events`, `/api/cache`, …) plus the bundled web UI (workflow graph, live runs, agent activity, cache inspe

Atomic Mail agentic core — PoW auth, JMAP, presets, and help for integration hosts.

Agentic Automation RunTime — a governed block/workflow runtime for AI agents

<div align="center"> <img src="https://raw.githubusercontent.com/revanthpobala/skill-atlas/main/public/logo.svg" alt="Skill Atlas Logo" width="120" /> <h1>Skill Atlas Workbench</h1> <a href="https://www.npmjs.com/package/agentic-skill-validator">

Agentic-checkout-ready address validation — converts Acuris /validate results into UCP (Universal Commerce Protocol) messages and ACP (Agentic Commerce Protocol) message errors inside merchant checkout-session endpoints.

OpenAI adapter using the official OpenAI Node SDK for agentic chat completions.

Autonomous development pipeline — aa: · ag: · audit: · AST graph · Harness · Specs · Impact analysis · Decision trail · Metrics · MCP server. Works with Cursor and Claude Code.

Agentic AI Development Life Cycle — specialized AI agents, enforced quality gates, and multi-LLM support for every phase of software delivery

Drop-in audit/compliance callback handlers for LangChain, LangGraph, CrewAI, AutoGen, Mastra, PydanticAI, and Microsoft Agent Framework. Streams events into the Ujex hash-chained audit ledger (IETF draft-sharif-agent-audit-trail-00). Resolves LangChain RF

LangGraph/LangChain framework wrapper for obs-unified Agent Action Graph

Zod v4-first LangGraph StateGraph wrapper with standardized node schemas, hooks, logging, checkpointing, test helpers, and resume support.

LangGraph.js checkpointer for PerSQL — durable agent state in an isolated SQLite database per agent

Apply Digital's AI coding assistant CLI — a Claude Enterprise harness for Apply Digital developers with BMAD support, multi-agent orchestration (Claude, Gemini, OpenCode), and secure Google SSO via Anthropic.

Papaya AI agent trace capture SDK for production apps.

AI-powered interview robot that conducts async multi-turn interviews via DingTalk

AgentOS terminal CLI — drive a real LLM agent on the microkernel; strongest-UX shell, advanced agent core

Node.js SDK for Croo — AI agent marketplace on Base

LangChain tools for Atlassian Jira and Confluence Cloud — direct REST calls, no MCP, proxy-aware. Extracted from Jarela.

LangChain tools for Atlassian Jira Align (portfolio / SAFe agile) — direct REST calls, no MCP, proxy-aware. Extracted from Jarela.

LangChain tools for the GitHub REST API — direct REST calls, no MCP, no gh CLI, proxy-aware. Extracted from Jarela.

n8n community node for the Google Gemini Chat Model, built on @intrvls/langchain-google-genai (the @google/genai-based port of @langchain/google-genai)

AgentMetrics observability integration for LangChain.js

Native LangChain BaseChatModel adapters for the Claude Code OAuth subscription (Anthropic) and opencode/Zen — no Pi, no API key.