AI agents

2,941 results · Page 41 of 82

cmd-mcp-server

CMD-MCP-Server is a server implementation based on the Model Context Protocol (MCP), used to execute CMD commands on Windows and Linux systems, and supports SSH connections. It allows command-line operations to be integrated with MCP-compatible applications (such as Claude), thereby extending agent capabilities. The project is written in TypeScript, based on the official MCP SDK, and has cross-platform compatibility. It can be automatically installed to Claude Desktop via Smithery, or installed manually (requires Node.js v16+ and npm/yarn). The server can be configured through environment variables or a configuration object. The project uses the MIT license. Note: executing command-line operations may pose risks, and security measures and input validation must be implemented in production environments.

cmdxray

Offline safety gate for agent shell commands: rates each command as danger, caution or none and explains every flag, pipe, redirect and subshell.

cmux

cmux is an open-source macOS terminal application based on the Ghostty rendering engine, built natively with Swift and AppKit, and designed specifically for running multiple AI coding agents simultaneously. It provides vertical tabs, notification rings, and a notification panel, making it easy to quickly locate the agents that need attention across multiple tasks and sessions. cmux is compatible with existing Ghostty configurations, supports GPU-accelerated rendering, a built-in programmable browser, a CLI, and a Unix socket API, and can create workspaces, split panes, send keystrokes, and automate browser operations. It supports session restoration, SSH remote workspaces, tmux/zellij local persistence, and notification and restoration integration for various agents such as Claude Code, Codex, and OpenCode. The software is free, GPL-licensed, and currently supports macOS only.

Cobot

Cobot is an AI agent collaboration tool developed by the Mainframe team, designed to enable multiple AI agents to collaborate on tasks in a shared space, where users can interact with the agents and complete work together. The product is currently in the process of being shut down: billing has been suspended as of the announcement date, annual subscribers will automatically receive prorated refunds, and all users can continue to have full access until November 30, 2025; after that, all user data will be securely deleted in accordance with the privacy policy. The official statement says the shutdown is to focus on a new core direction, but no specific details were disclosed. Cobot was previously in an early access beta phase, and details such as its specific features, supported platforms, and pricing have not yet been provided. For account issues, you can contact help@cobot.co.

Cocreator AI

creative computer agent (in wait list)

Code Autopilot

The introduction is being prepared. Please check back soon.

code-assistant

Coding agent with basic tools for listing, reading, editing and writing files, running commands and web search, supporting multiple projects at once.

code-cli

Autohand Code CLI is a fast, self-improving AI coding agent that runs in the terminal, developed by autohandai. It reads project context, plans changes, edits files, calls tools, and requests approval before performing high-risk operations. It supports an interactive REPL and non-interactive command mode for CI/CD, with built-in file editing, Git integration, planning and memory, 40+ tools, modular skills, and compatibility with multiple providers such as OpenRouter, OpenAI, AWS Bedrock, DeepSeek, Azure Foundry Models, Z.ai, and local models. It runs on macOS, Linux, and Windows, and offers editor extensions and multilingual SDKs.

code-relay

Branch-scoped runbooks and cross-machine verification with auditable receipts for coding agents.

CodeCortex

Persistent codebase knowledge for coding agents: symbols, dependency graphs, co-change patterns and architectural decisions extracted with tree-sitter.

codefuse-chatbot

CodeFuse-ChatBot is an open-source AI intelligent assistant developed by Ant's CodeFuse team, designed to serve the software development lifecycle. It is based on a Multi-Agent collaborative scheduling framework and integrates a tool library, code repository, knowledge base, and sandbox environment. It supports retrieval-augmented generation (RAG), tool learning, and sandbox execution, helping to handle complex tasks in the DevOps field. The project supports offline private deployment of open-source LLMs and Embedding models, and also supports OpenAI API calls.

CodeGraphContext

Index local code into a graph database to give AI assistants context, with graphical code visualizations for humans.

codemcp

Coding agent with basic read, write and command line tools.

Coder

Self-hosted cloud development environment for developers and AI coding agents.

codex-control-plane-mcp

Control plane for long-running Codex Desktop tasks: submit asynchronously, poll state, approve Plan Mode, recover retries and search SQLite-backed history.

codex-mcp-tool

Connect an IDE or AI assistant to Codex CLI for code analysis and editing, with support for multiple models.

codex-plugin-cc

Claude Code plugin that enables using OpenAI Codex for code reviews and task delegation directly within Claude Code workflow. Provides /codex:review, /codex:adversarial-review and other commands

Cody by Sourcegraph

Cody is an AI coding assistant from Sourcegraph, built on the latest LLMs and combined with the user's development context to help understand, write, and fix code. It uses Sourcegraph's advanced search API to extract context from local and remote code repositories, so it can draw on context about APIs, symbols, and usage patterns from across the entire codebase. Cody seamlessly connects to code hosting platforms such as GitHub and GitLab, as well as IDEs such as VS Code, JetBrains, and Visual Studio, and serves as a personal AI coding assistant offering capabilities such as developer chat, code completion, code editing, and customizable prompts. It is primarily aimed at teams and individual developers using Sourcegraph Enterprise, and supports multiple ways of use.

cognirepo

Persistent memory and codebase context for coding agents: FAISS vector store, AST symbol index, knowledge graph, episodic journal and cross-agent handoff.

Coin wizard

AI-powered Bitcoin trading automation tool.

coinopai-mcp

Paid crypto intelligence via x402 on Base: preflight checks, trade decisions audited against real prices, risk state, signal history and automation search.

coinrithm-agent-trading

Paper-trade crypto spot, futures and prediction markets on CoinRithm with simulated funds and a public Agent Arena.

ColaOS

The first AI Agent operating system with a soul

coldstart

Codebase memory for coding agents: a deterministic AST index finds task-relevant files, and in-repo Markdown notes flag themselves stale when code changes.

comet-mcp

Connect to the Perplexity Comet browser for agentic browsing, deep research and real-time task monitoring.

Comp AI

Comp AI is an open-core compliance platform (AGPL-3.0) that uses AI agents to collect audit evidence, capture screenshots from vendor apps, draft policies, monitor cloud and device settings and run AI penetration tests for frameworks such as SOC 2, ISO 27001, HIPAA and GDPR. It suits startups and mi

Compliantchatgpt

HIPAA-compliant AI agent for healthcare, using OpenAI with data security.

ComplyEdge

Runtime compliance enforcement layer for AI agents that checks every prompt or response against legal rule corpora and returns article citations, rule IDs, timestamps, and block decisions instead of a generic safety score.

Composable prompts

Low-code GenAI platform for rapid development, deployment, and operation of AI applications and agents.

Compozy

CompozyOS is a complete runtime environment built around AI agents, designed to let agents work continuously without having to write their own scripts, cron jobs, or glue code. It treats loops, automation, permissions, approvals, and run history as core objects within the same runtime, accessible via CLI, HTTP, and UDS. At its core is a true autonomy kernel: tasks atomically claim runs via ClaimNextRun, hold a lease and must send heartbeats. After a crash, the lease expires according to its TTL and the run is automatically re-enqueued, thereby avoiding duplicate execution and orphaned tasks. Humans and agents share the same queue (the task_runs table), with the same primitives and audit trail. Permissions are enforced in code through generated contracts, and a child cannot expand a parent's permissions. It also provides persistent sessions, observability tracing, Cron and Webhook automation, and pluggable layers such as hooks, skills, and extensions, with everything loaded being inspectable. It is currently in beta; installation options include a verified installer for macOS/Linux, the npm beta channel, and pinned Go distributions. The experience leans toward technical users: it requires a terminal, a local daemon, and a configuration file.

computer

Cloudflare Computer is an open-source virtual file system project from Cloudflare that runs inside a Durable Object, uses SQLite to store authoritative state, and provides pluggable execution backends through workspace.runtime, giving AI agents a file system and execution environment.

concord-mcp

Communication and shared work state across coding agent harnesses: live messages, scope claims before editing, preserved decisions and evidence handoffs.

Conductor

Conductor is a platform for running coding agents in the cloud. Through isolated cloud sandboxes, it lets users run multiple coding agents simultaneously and supports multiplayer collaboration. Conductor Cloud is based on isolated microVMs, comes preloaded with repositories and dependencies, sandboxes can start in seconds, and agents can run for long periods. The platform has built-in first-party agents such as Claude Code, Codex, Cursor, and OpenCode, and allows users to bring their own subscriptions and keys. Users can operate it through the desktop app, mobile, or their own integrations (via the Conductor API), and it supports sharing workspace links and real-time collaboration with teammates. The company says more than 100,000 builders already use it.

conductor

Netflix-originated open-source durable workflow engine now focused on AI agent orchestration. Provides event-driven execution, durable state, retries/timeouts, 14+ LLM providers, MCP tool calling, function calling, human-in-the-loop approval, and vector DB integration for RAG. Trusted in production

conduit8

The introduction is being prepared. Please check back soon.

Connapse

#️⃣ 🏠 🍎 🪟 🐧 - Self-hosted knowledge backend for agents with hybrid vector and keyword search and container-isolated indexes.