Code agents

🛡️ Experimental security guardrails for LangChain agent code execution — Docker sandbox isolation.

Unofficial smolagents Model adapter for ChatGPT Codex OAuth device-code auth.

A strict Spec Agent workflow for spec-driven development

Claude Code (claude_agent_sdk) DTAP agent target for superred: subclasses dtap_scaffold.DtapAgentTarget and runs the Claude Agent SDK in an isolated Docker container against the DTAP env MCP proxy, normalizing the in-container transcript into the shared TrajectoryArtifact.

Graph-first multi-agent orchestrator for Claude Code - parallel task graphs, hook-written checkpoints, ask-first resume, right-sized model tiers

Local coding agent built on LangChain Deep Agents (LocalShell, no sandbox)

LLM agent with Markdown config, tool loop, Web UI, sessions, memory, and webhooks

Ready-made agents for the TAI ecosystem — coding/deep agents, tool-calling, retrieval, voting, refinement, and vision agents — opt-in and manifest-loaded.

Self-hosted event server that pushes a notification the instant a parallel coding-agent session blocks, needs review, fails, or finishes.

Local execution daemon for Tether Brain agent installations

Anthropic integration for telemetry.dev Python SDK

AWS Bedrock integration for telemetry.dev Python SDK

LiteLLM integration for telemetry.dev Python SDK

OpenAI integration for telemetry.dev Python SDK

A secure, provider-independent coding agent for the terminal

Repository-owned agent skills that just work

Local HTTP control daemon for tmux-backed coding-agent sessions

Pi-style LM authentication helpers for DSPy

Shared memory and skills across Claude Code and Codex

Universal Repository Agent Graph: structure-aware, token-efficient RAG for projects

Local-first agentic trading AI agent with on-device Pilot Router

A maintained successor to user-agents: identify devices (phones, tablets) and their capabilities by parsing browser user agent strings.

A veto gate for AI-written code in LangChain / LangGraph — 17 reality gates + frontier review, with veto power. Free to try, no signup.

LangChain integration for viparse — read Vietnamese documents, legacy encodings included

LlamaIndex integration for viparse — read Vietnamese documents, legacy encodings included

A minimal, transparent, local-first Waku: harness + loop + memory + eval, in code you can read in an afternoon.

Architectural memory system for autonomous coding agents.

Harness-agnostic adapter that traces coding-agent sessions to W&B Weave

Ton année Claude Code, wrappée en néo-brutalisme (CLI). Adaptation de My Claude Wrapped (@HKafuiEPI).

Claude Agent SDK–shaped Python API for the Grok Build harness — any model

小羽 — a harness coding agent

Modular Transformer & Multi-Corpus (English, Arabic, Code) LLM Library

Bridge non-ADMS/non-Cloud-Server ZKTeco devices to application's ADMS HTTP (attendance, commands, enroll)

A sleek and professional agentic coding console and CLI assistant.

Rierino review: compare features, pricing, use cases, access model, and alternatives for this AI Agents Platform agent in 2026. Low-code platform to build, deploy, and govern enterprise AI agents that execute real actions across your systems.

Capy QR review: compare features, pricing, use cases, access model, and alternatives for this Other agent in 2026. Free QR code generator with logos, colors, and borders

Laravel/PHP AI agent skill — compatible with Claude Code, Cursor, Windsurf

Laravel/PHP AI agent skill — compatible with Claude Code, Cursor, Windsurf

Laravel/PHP AI agent skill — compatible with Claude Code, Cursor, Windsurf

Governed AI agent orchestration with repository intelligence, approval gates, specialist workcells, and reviewable evidence.

Two AI agent seats in one window: one builds, one reviews.

AI agent integration plugin for Baseplate — generates AGENTS.md, CLAUDE.md, .mcp.json, and .agents/ configuration files

Local-first GitHub-backed sync for AI agent skills, instructions, and Codex plugins.

轻灵 Qling - 本地优先的中文 AI Agent CLI 工作台 / Local-first Chinese AI Agent CLI workbench

Provider-independent AI agent workflow system for Claude Code, Codex, Gemini CLI, and Kimi Code

AI Agent skill with prompting set for development process

TikTok CLI - designed for AI Agent and developers

Local-first archive, search, and recall for AI-agent conversations.

Move your AI agent between clients — config, MCP servers, skills, memory, persona. Any direction, dry-run first, lossless where possible.

SOLAI Agent is a local-first AI agent and coder that runs locally on your computer.

JS reverse engineering MCP server with agent-first tool design and built-in anti-detection. 为 AI Agent 设计的 JS 逆向 MCP Server，内置反检测。

Vision Hub CLI - query projects and resources, designed for AI Agent and developers

Brain-inspired associative memory layer for an LLM agent. Automatic (implicit) tier: capture -> embed -> recall. Codename Factor-X.

Dependency-free static analyzer that audits LLM agent tool/function definitions (OpenAI, Anthropic, MCP) for excessive agency, code-execution & SSRF sinks, secret access, and data-exfiltration surface — mapped to the OWASP 2025 LLM Top-10. Runs in CI.

Usage tracking, credit management, execution-trace collection, and telemetry export for the Agent Dev Container SDK.

AgentTap — agent-native, local-first LLM observability in one process: four capture channels (OTLP traces/logs ingest, built-in LLM tracing proxy, Claude Code transcript import with input reconstruction), SQLite store, and a LangSmith-style three-pane das

Define LLM-powered backend actions as Markdown. Generate typed clients. Run them anywhere.

Vendor-neutral control-plane contract for coding agents: a normalized ToolCallEvent/Verdict schema with per-agent adapters (Claude Code, Codex, Amp, Gemini CLI) and a conformance suite.

Defend an agent against hidden-content injection: strip payload-capable invisible Unicode and ANSI, splice out human-invisible HTML, and flag data-exfil URLs in untrusted text before any model sees it.

AI-powered code review from your terminal — security, performance, and quality agents on every PR

A TypeScript CodeAgent harness reconstructed from Arena message and tool traces.

A small, provider-neutral TypeScript toolkit for calling LLMs from backend code.

AI-powered vulnerability scanner for any codebase

Cross-platform OpenAI-compatible Qwen coding agent CLI

A lightweight coding agent with OpenAI-compatible and Anthropic Messages API support.

Universal context-compression proxy for AI coding agents. Sits between any agent and its model API, rewriting Anthropic/OpenAI streams with acp-kernel compression. Any agent that can set a base URL (Claude Code, Codex, Cursor, Aider) works out of the box.

Turn a supported ChatGPT/Codex OAuth sign-in into a private OpenAI-compatible endpoint, starting with self-hosted n8n.

Compose and run coordinated subagent staff packs with progressive discovery, portable Markdown personas, and no runtime or dependencies.

OpenAI fast mode toggle for pi - Enables priority service tier on supported GPT-5 models

Local-only agent memory store CLI

Vercel AI SDK provider adapter for openai-oauth auth handles.

Universal AI coding-agent runtime and model-routing layer for Claude, Codex, AGY, and compatible runtimes.

Local Codex auth-file auth handle for openai-oauth.

[](https://www.npmjs.com/package/@agentclientprotocol/codex-acp)

Swiftx — Native CLI coding agent

Kimi Code models in pi-coding-agent with OAuth, K3, K2.7, and HighSpeed selection

VAIA SDK — platform integration for Gandia-7 and Handeia developers

Autonomous Multi-Agent Orchestrator & Ecosystem of 44+ specialized AI experts for Antigravity, Cursor, Claude Code, Windsurf, Cline, Roo Code, and Copilot.

Powerful AI coding agent CLI with rich UI, project memory, and smart tools.

Forklane — Autonomous Coding Agent. Terminal-based AI coding agent with multi-provider LLM support, four-tier memory, verification pipeline, Forklane SSH remote execution, and collaborative sessions.

OpenCode-style default agent mode for PI. Markdown agents with YAML frontmatter. Ctrl+Shift+M to cycle, Alt+S to search. Inline execution with full streaming visibility.

CLI to initialize an agent OS: CLAUDE.md protocols, committed .ai/ knowledge layer with autonomous codebase learning, modes, and git-driven staleness detection

An autonomous coding agent for building AI-powered development tools. The name "Hataraku" (働く) means "to work" in Japanese.

ESLint plugin for Vercel AI SDK security — detects prompt injection, system-prompt leaks, hardcoded API keys, and unvalidated model output in generateText and streamText.

Pi extension for single-agent delegation and scripted multi-agent workflows

Open harness for reproducible multi-agent coding workflows

Just A Rather Very Intelligent System - A developer-first, type-safe, transparent multi-agent SDK

Multi-agent BMAD framework for Claude Code — SPEC → DESIGN → BUILD → QA in one kit.

Multi-agent operating model for Claude Code and Cursor — roles, contracts, channels, and script gates

Multi-agent coding runtime with a real-time collaboration WebUI, durable tasks, CLI, and SDK

CLI-based multi-agent coding orchestrator — plan, execute, review on top of any coding agent

Multi-agent compliance skills for open-source GRC (Claude Code, Codex, OpenCode) powered by Probo MCP

Conversation memory skill for multi-agent AI workflows — store, share, and recall across sessions and agents

Multi-Agent Orchestration for AI Coding Assistants — task management, git isolation, file locking, and real-time dashboard

TaskFlow MCP server + Claude Code hooks — multi-agent task board, realtime chat, reviews, terminal streaming, and activity logging for AI coding agents.

Zero-dependency multi-agent orchestrator with a file-based mail model for AI coding agents (Claude Code, Codex, Gemini, Hermes). Agents talk by reading and writing files; the orchestrator routes the mail.

MCP Server for Spec-Driven Development (SDD) — contract validation, task board, and file reservation for multi-agent AI systems

Spec-driven development toolkit for AI coding agents: end-to-end workflows, multi-agent coding and review, living specs, and engineering guardrails

Auditable multi-agent coordination for OpenCode — parallel agent fleets with P2P messaging, witnessed receipts, and quorum ratification. Who saw this, who approved it, prove it.

AI Coding Agent Framework — interactive bootstrap CLI