npm agents

Core engine for AI agent — LLM orchestration, tool execution, MCP, skill routing, security, and memory pipeline.

MCP server for AgentPlanner — AI agent orchestration with planning, dependencies, knowledge graphs, and human oversight

Constella — run autonomous AI agent-companies locally: real claude/codex agents, Goals → Specs → Issues → Plans, local models, GitHub & Telegram, all from one cosmic control plane.

The official JS/TS SDK for Creduent: the open standard and security protocol for cryptographic AI agent identity, authentication, and trust verification using Ed25519 and DNS.

MCP server for SpecForge - AI agent integration

Zero-to-production AI agent deployment framework

SpendGuard pre-call budget enforcement for n8n AI Agent workflows — drop a sub-node between Chat Model and AI Agent.

AI-agent-friendly CLI for controlling LinkedIn accounts and retrieving real-time data.

AI agent rules for Contentrain — quality standards, architecture conventions, IDE integration

This package is a thin OpenAPI wrapper for Nebula, our AI agent service.

WasmAgent Compliance Engine — TaskSpec-driven verification, local repair, and evidence export for LLM agent runs.

Tiered, token-budget-aware memory for any LLM agent

MCP server for Plan Budowlany — manage construction project tasks, costs, issues and timeline from an LLM agent.

MCP server adapter for Kiba — connects any LLM agent to the marketplace via OAuth

A minimal, opinionated set of coding tools for driving an LLM agent over a workspace — read/write/edit/patch/move, batch read, grep/glob/diff, project-wide regex replace, bash + background monitors, and tree-sitter-powered outline/check_syntax — usable as

Fetch an OpenTelemetry trace from Axiom or Jaeger and render it into a single, deterministic Markdown file — structured for human reading and LLM agent consumption.

Visual node-editor for tramo workflows — React components and LLM-agent glue. Pairs with @tramo/spec (the wire format) and @tramo/runtime (the executor).

Workflow examples for Human-in-the-Loop patterns in Loopstack — custom Document with widget, AskUserWorkflow / ConfirmUserWorkflow sub-workflow shortcuts, LLM agent loops with ask_clarification / ask_for_approval, and a meeting-notes review flow.

Provider-agnostic token profiler for LLM agent loops. See exactly where your context tokens go — and how many you waste re-sending the same thing.

AI red-teaming for the AI agents & LLM apps you own: stress-test them with adversarial models to find security failures (prompt injection, tool misuse / excessive agency, data leakage, jailbreaks, denial-of-wallet), auto-patch (blue team), retest, and get

Live, agent-driven Refine panel for CSS/Motion transitions — injects a timeline + Refine UI and runs transitions.dev suggestions via your coding agent.

二郎神 - 全知全能的AI投资智能体

Universal AI agent runtime — portable .agent files run identically in cloud, desktop, CLI, and as an MCP server. Multi-LLM (Anthropic, OpenAI, Gemini), pluggable adapters, MCP client + server, streaming, abort, pause/resume.

General-purpose agent with transport abstraction, state management, and attachment support

Opt-in Langfuse trace sink for the Agent Architecture Board. A fetch-based LlmTraceSink that streams LLM generations (and container tool spans) to Langfuse's ingestion API — runs unchanged on both the Cloudflare Worker (workerd) and Node facades.

General-purpose agent with transport abstraction, state management, and attachment support

Knight-agent智能体架构

Agent evaluation and benchmarking for AgentsKit.

Streaming colorization for Shiki, useful for highlighting LLM outputs and other text streams

TrimPrompt — Save 80% on LLM tokens. Zero-config CLI proxy for AI coding agents

Pane-embeddable coding agent. Ships direct/local/vercel-sandbox execution modes behind one interface.

Use LLMs to robustly extract and enrich structured data from HTML and markdown

Portable hybrid code auditing + remediation orchestrators for arbitrary repositories.

Typebulb CLI to run single-file markdown apps called bulbs, either as standalone web apps or embedded in agent responses.

own your computer-use — your LLM on your own mouse and keyboard. Part of Own Your Stack.

Local AI coding agent for your terminal — an open-source, offline alternative to Claude Code, Cursor, and GitHub Copilot, powered by Ollama. Private by default, free forever.

AI-native Azure DevOps CLI — structured JSON output, embedded skills for LLM agents, single-file cross-platform binary.

General-purpose agent with transport abstraction, state management, and attachment support

LangChain.js / LangGraph callback adapter for AgentInspect local execution trees and Evidence workflows

Paybond Kit for TypeScript: agent spend governance for paid tool calls with spend authorization, evidence receipts, refunds, disputes, hosted Gateway sessions, and settlement.

Agent-callable wrappers for Veil Protocol's quantum-resistant payments (x402-pqc) and private Ghost AI inference (circles) — MCP server, LangChain tools, OpenAI functions, and a REST API

On-Device Prompt API Chain AI Agents library

LangChain tool adapter for the read-only BRADY Discovery SDK — advisory routing and observation tools for agent workflows

LangChain TS callback handler for SpendGuard budget guardrails

Slack adapter for Chat SDK — webhooks, Block Kit cards, OAuth, slash commands, and AI streaming

LangChain integration for the Grantex delegated authorization protocol

LangSmith tracing module and decorators for NestJS.

Render markdown to Telegram, stream agent replies into live drafts, and drive a snapshot/rollback turn-loop — runtime-agnostic.

NestJS decorators and discovery for LangChain tools.

Human-in-the-loop MCP server for AI agents — biometric approval with PKI-signed attestation. stdio + HTTP.

Enprompta framework instrumentation for LangChain — emits typed, nested OpenTelemetry spans that Enprompta ingests as first-class Retrieval/Tool/Agent spans.

Telegram adapter for Chat SDK — Bot API webhooks, inline keyboards, reactions, and streaming

Linear adapter for Chat SDK — @mentions in issue comments and Linear agent sessions

> AI framework integrations for the Yuno payment orchestrator

Thin NestJS registry and shared contracts for optional LangChain integrations.

PayDuct spend-request tool for LangChain agents

Local AI executor with a CLI, local API server, and web UI.

One install for Stratos AI: OpenLLMetry agent telemetry (--import) AND the stratos-mcp-proxy for MCP servers.

Runtime adapter that bridges agent frameworks (OpenAI SDK, LangChain) with production infrastructure

Vercel AI SDK telemetry integration for AgentInspect local traces and trajectory tests

A [HarnessV1](../harness) adapter that runs [Deep Agents](https://github.com/langchain-ai/deepagentsjs) (LangChain's LangGraph-based agent harness) as a coding-agent runtime inside an AI SDK sandbox.

NestJS prompt registry helpers for LangChain prompt templates.

LangChain tools for PayDirect — enable AI agents to make payments, check balances, swap tokens, and manage wallets

OpenAI-compatible provider tokens and model factory wiring for Nest LangChain.

CLI for onboarding and managing target repositories.

Gemini provider tokens and model factory wiring for Nest LangChain.

Hosted graph documentation and layout APIs for Nest LangChain registries.

Microsoft Teams adapter for Chat SDK — Bot Framework webhooks, Adaptive Cards, and Graph API

Anthropic provider tokens and model factory wiring for Nest LangChain.

Discord adapter for Chat SDK — HTTP interactions, Gateway events, slash commands, and embeds

Unbrowse as native LangChain JS tools. Three tools (resolve, execute, search) shaped exactly like LangChain's StructuredTool/DynamicStructuredTool — { name, description, schema (JSON Schema), invoke, call }. No zod dependency. Pass your own @langchain/cor

Anchor your LangChain.js agent's run tree as it executes: one callback handler turns every chain/LLM/tool step into a Merkle-batched, individually inclusion-proofed provenance record on ar.io — prompts and outputs are hashed locally and never uploaded.

Paperclip adapters for Hermes Agent local CLI and Hermes Gateway HTTP/SSE runs

Friday coding agent CLI/TUI app.

Deterministic model recommendation CLI for OpenCode OMO agent configurations

OpenAI Agents JS processor integration for AgentInspect local traces and TraceContract checks

OpenClaw IM channel gateway — bridge Telegram/Discord/WeChat/DingTalk/QQ to any OpenAI-compatible agent API

A declarative, verifiable graph of task nodes for the Pi coding agent — statically verified before it runs, with dynamic fan-out, gates, isolated subagent context, resumable runs, and saveable commands.

MCP vision bridge for text-only coding agents

Agent observability and knowledge capture layer for AI coding tools.

OpenCode plugin for LLM prompt caching, token savings, and KV cache reuse across AI agent sessions

UR-Nexus — autonomous engineering workflow engine (plan, execute, test, verify, document, benchmark, reproduce)

n8n node for OpenAI GPT-5 with PDF processing, web search, and timeout configuration via Responses API

A Node.js library for various integrations including payments and LLM services - exact port of the Python emergentintegrations package.

An AI coding agent that lives in your terminal — it reads, edits, and runs your code to finish real tasks, asking before it changes anything. Start with one Nexel access key (no model setup) or bring your own OpenAI, Anthropic, or Azure key. Streaming rep

Standalone, browser-compatible SDK providing Claude Code agent capabilities (tools, tool loop, multi-turn, MCP, sub-agents, sessions) against any OpenAI/Anthropic-compatible LLM endpoint. Runs in the browser (WebContainer), Node, and Bun — no backend requ

Self-hosted n8n Nils Codex node via ChatGPT subscription device-code auth

Standalone mixdog coding-agent CLI/TUI workspace.

Polls GoNext cloud API for async local LLM jobs and runs them against Ollama/OpenAI-compatible servers on this Mac

Custom Provider CLI for managing model providers

Local privacy gateway — redacts PII out of OpenAI/Anthropic API traffic before it reaches a model, then restores it in the reply. Point opencode, codex, aider, Claude Code, etc. at it.

API pour l'orchestration d'agents intelligents avec séquences et escalades automatiques

OpenAI-compatible embedder for ObjectStack — works against OpenAI, 阿里通义 DashScope, 智谱 BigModel, 硅基流动 SiliconFlow, 火山引擎 Doubao, MiniMax, Ollama, and any drop-in OpenAI-shape endpoint.

Orionfold Relay — a local-first, multi-agent orchestration runtime and builder scaffold for AI-native work.

Embedded memory for TypeScript agents. Local-first by default. Pluggable to edge and serverless. Extract facts, embed them, and recall scoped memories from your app process, with pluggable vector storage and framework adapters.

Local-first coding agent memory for Claude Code and OpenAI Codex

OpenAI / GPT provider for chat completions.

Dynamic MCP servers + agent skills as uniform tools for any LLM (opencode-style).

[](https://deepwiki.com/madebywild/agent-harness)

Drop wasmagent sandbox kernels into the OpenAI Agents JS SDK as a Tool — edge-safe code execution with one capability manifest, no E2B sandbox needed