* [oh-my-claudecode v5.0.0](https://github.com/Yeachan-Heo/oh-my-claudecode) – Multi-agent orchestration for Claude Code with autonomous, parallel, coordinated, pipeline and token-efficient execution modes, specialized agents, and real-time HUD visibility.

* [Claude-Mem v13.16.0](https://github.com/thedotmack/claude-mem) – Persistent memory compression system that preserves and summarizes context across Claude Code sessions for continuity.

* [Gemini CLI v0.57.0](https://github.com/google-gemini/gemini-cli) – Command-line AI workflow tool that connects to tools, understands code, and accelerates workflows.

* [NanoClaw v2.3.0](https://github.com/nanocoai/nanoclaw) – Lightweight AI assistant running Claude agents in isolated containers, with messaging integrations, memory, scheduled jobs, and Anthropic Agents SDK support.

* [RAGFlow v0.27.0](https://github.com/infiniflow/ragflow) – RAG engine combining large language models with document understanding for accurate question answering and citation.

* [OneCLI v2.0.0](https://github.com/onecli/onecli) – Gateway that injects stored credentials into outbound agent HTTP requests, swapping placeholders for real keys so agents never see secrets.

* [AnythingLLM v1.16.0](https://github.com/Mintplex-Labs/anything-llm) – Full-stack application for turning documents into context for LLMs with multi-user management.

* [OpenCode Swarm v7.151.0](https://github.com/ZaxbyHub/opencode-swarm) – Architect-led agent swarm plugin for OpenCode that runs gated code generation, review, testing, security checks, and documentation without parallel mutations.

* [Open Mercato v0.7.0](https://github.com/open-mercato/open-mercato) – AI-supportive, modular framework for building enterprise-grade CRM, ERP, and commerce backends with customizable modules and multi-tenant support.

* [DSH-IM v3.0.0](https://github.com/xmanrui/dsh-im) – DeepSeek Harness plugin that connects IM bots via QR code or credentials across multiple chat platforms.

* [Open Design open-design-v0.21.0](https://github.com/nexu-io/open-design) – Local-first, web-deployable design workflow that turns your coding agent into an artifact-driven design engine with composable skills and design systems.

* [UX/UI Agent Skills v3.0.0](https://github.com/plugin87/ux-ui-agent-skills) – Instructional kit and runnable skills that enable Claude to generate token-driven, accessible components and design systems for any framework.

* [J-Space Cognition Suite V3.7 v3.7.0](https://github.com/Tiger3807861189/J-Space-Cognition-Suite-V3.7) – Model-agnostic inference-time control suite that manages a workspace for deep reasoning, long-horizon work, tool use, verification, and recovery through selectively loaded skills.

* [sem v0.23.0](https://github.com/Ataraxy-Labs/sem) – Semantic version control providing entity-level diffs, blame, and impact analysis via tree-sitter across ten programming languages.

* [Big-AGI v2.1.0](https://github.com/enricoros/big-AGI) – Multi-model AI workspace for experts offering advanced chat, persona controls, multi-model validation, and local-first deployment.

* [Rowboat v0.9.0](https://github.com/rowboatlabs/rowboat) – Local-first AI coworker that builds persistent, editable Markdown knowledge from emails and meeting notes to assist everyday work.

* [Watchfire v10.0.0](https://github.com/watchfire-io/watchfire) – Orchestrates AI coding agents with project-aware context, structured tasks, and isolated git worktrees for iterative code generation.

* [Pipelock v3.4.0](https://github.com/luckyPipewrench/pipelock) – All-in-one security harness for AI agents offering egress control, DLP, SSRF protection, prompt-injection scanning, and workspace integrity monitoring.

* [vibe-coding-prompt-template v3.1.0](https://github.com/KhazP/vibe-coding-prompt-template) – Practical AI workflow template that turns ideas into MVPs with structured prompts, research, PRDs, tech designs, and agent docs scaffolding.

* [OpenConnector v1.4.0](https://github.com/oomol-lab/open-connector) – A connector gateway that lets AI agents and apps access 1,000+ SaaS providers via a shared catalog of 9,400+ prebuilt actions.

* [CLI Printing Press v4.31.0](https://github.com/mvanhorn/cli-printing-press) – Prints token-efficient, agent-focused Go CLIs with Claude Code skills and MCP servers by extracting APIs and syncing data to SQLite.

* [StashBase v2.0.0](https://github.com/liliu-z/stashbase) – Indexes and converts local files into searchable, agent-ready text with semantic and keyword search via MCP clients.

* [Open Multi-Agent v1.16.0](https://github.com/open-multi-agent/open-multi-agent) – TypeScript-native multi-agent orchestration that decomposes goals into parallel task DAGs and synthesizes results.

* [Glowbom OSS v4.0.0](https://github.com/glowbom/glowbom-oss) – Coding-agent workflow that helps turn real software projects and prototypes into production-ready code using locally configured OpenCode providers and models.

* [okf-gem v2.0.0](https://github.com/serradura/okf-gem) – Open Knowledge Format harness providing an agent skill, Ruby CLI/library for validation and search, and graph server or render for managing Markdown knowledge bundles locally.

* [agentcookie v1.0.0](https://github.com/mvanhorn/agentcookie) – Continuously sync Chrome cookies, CLI secrets, and auth blobs from your primary Mac to a secondary Mac over an encrypted Tailscale link so headless agents wake up authenticated.

* [ArcKit v6.9.0](https://github.com/tractorjuice/arc-kit) – ArcKit is a toolkit for enterprise architects that turns architecture governance into a structured, AI-assisted strategy, delivery, and assurance workflow.

* [@tintinweb/pi-subagents v0.19.0](https://github.com/tintinweb/pi-subagents) – pi extension bringing Claude Code-style autonomous sub-agents with parallel background execution, live widget, custom agent types, and mid-run steering.

* [Pneuma Skills v3.43.0](https://github.com/pandazki/pneuma-skills) – Co-creation infrastructure connecting humans and code agents with visual environments, skills, continuous learning, and distribution.

* [lnwjud v4.13.0](https://github.com/engasnm111/lnwjud) – Windows-first local AI-agent runtime and MCP gateway that exposes trusted capabilities through MCP for files, Git, processes, Windows automation, WSL, browser control, indexing, and observability.

* [Claude Code Workflows v0.25.0](https://github.com/shinpr/claude-code-workflows) – End-to-end AI-assisted development workflows using specialized agents to plan, implement, and quality-check reviewable code.

* [wmux v3.48.0](https://github.com/openwong2kim/wmux) – Native Windows terminal multiplexer with split panes, MCP browser automation, and multi-agent AI CLI support without WSL.

* [Kortix v0.13.6](https://github.com/kortix-ai/suna) – Autonomous company OS where Linux-sandboxed AI agents run code, manage files, and automate operations 24/7.

* [OmniRoute v3.8.50](https://github.com/diegosouzapw/OmniRoute) – AI gateway providing an OpenAI-compatible endpoint with multi-provider routing, load balancing, retries, rate limits, caching, and observability for cost-aware inference.

* [FlowBoard v5.2.0](https://github.com/rasimme/FlowBoard) – Project workspaces for AI agents providing context management, agent-native Kanban, specs, and an idea canvas.

* [Arcjet v1.11.0](https://github.com/arcjet/arcjet-js) – SDK for implementing security features like bot detection, rate limiting, and email validation in JavaScript applications.

* [GrayMatter v0.15.0](https://github.com/angelnicolasc/graymatter) – Three-line embeddable agent memory layer providing persistent memory and reducing token usage by about 90%.

* [Collie v0.35.0](https://github.com/AltanS/collie) – Tailnet-served PWA web UI for a Herdr agent herd, showing the waiting agent and enabling reply via phone keyboard with push notifications.

* [MCPProxy v0.61.0](https://github.com/smart-mcp-proxy/mcpproxy-go) – Desktop proxy enabling scalable tool discovery, major token savings, and quarantine of malicious upstream servers for AI agents.

* [Ongrid v0.14.0](https://github.com/ongridio/ongrid) – AI agent that diagnoses, investigates, and fixes infrastructure issues via chat apps.

* [only-cli v0.5.0](https://github.com/only-cli/oc) – Web-browsing command-line tool for AI agents that fetches pages, renders compact numbered views, and enables token-efficient navigation and reading.

* [tossinvest-cli v0.43.0](https://github.com/JungHoonGhae/tossinvest-cli) – Command-line tool automating Toss Securities web workflows, offering account, quote, portfolio, watchlist, and read-only trading commands.

* [Activepieces 0.87.1](https://github.com/activepieces/activepieces) – Extensible AI automation platform with a type-safe TypeScript framework and over 280 integrations for workflow building.

* [agentglass v0.12.0](https://github.com/SirAllap/agentglass) – Mission-Control style dashboard and workspace that monitors AI coding agent activity in real time, with cost tracking, diffs, terminal, and source-control controls.

* [Memmy v1.1.0](https://github.com/MemTensor/memmy-agent) – Memory hub and local agent runtime that provides shared long-term personal context across multiple AI agents.

* [Veritas Kanban v6.1.2](https://github.com/BradGroux/veritas-kanban) – Local-first Kanban and AI-agent orchestration platform for developers to manage board-driven tasks, git worktrees, and autonomous coding workflows.

* [CyberStrikeAI v1.7.17](https://github.com/Ed1s0nZ/CyberStrikeAI) – AI-native, Go-based penetration-testing copilot combining hundreds of security tools, MCP orchestration, and an agent that runs engagements via conversation.

* [dev-3.0 v1.48.0](https://github.com/h0x91b/dev-3.0) – Kanban-first mission control that runs parallel AI coding agents using per-task git worktrees and tmux terminals to keep one workflow focused.

* [OCTO Server v1.16.0](https://github.com/Mininglamp-OSS/octo-server) – Go backend exposing REST and WebSocket APIs, orchestrating Lobster agent sessions, and running a WuKongIM control-plane for real-time messaging.

* [NVIDIA NemoClaw v0.0.114](https://github.com/NVIDIA/NemoClaw) – Reference stack simplifying secure, always-on OpenClaw assistant execution on the NVIDIA OpenShell runtime.

* [Obsidian Mind v8.3.6](https://github.com/breferrari/obsidian-mind) – Obsidian vault template that stores persistent, self-organizing memory for AI coding agents across sessions, with hooks for Claude Code, Codex CLI, and Gemini CLI.

* [Knowns v0.31.0](https://github.com/knowns-dev/knowns) – Local-first, self-hostable project context layer that stores structured tasks, docs, specs, and decisions and exposes them to AI assistants via CLI and MCP.

* [OpenCLI v1.8.7](https://github.com/jackwener/OpenCLI) – Universal CLI hub and AI-native runtime that turns websites, Electron apps, and local binaries into deterministic, scriptable command-line interfaces.

* [Agent Deck v1.15.0](https://github.com/asheshgoplani/agent-deck) – Terminal-based session manager for running, organizing, and switching between multiple AI coding agent sessions.

* [repo-harness v0.17.0](https://github.com/Ancienttwo/repo-harness) – A CLI and hook system that turns Claude/Codex coding into a repeatable, repo-local workflow with resumable plans, checks, and review evidence.

* [img2threejs v1.5.1](https://github.com/img2threejs/img2threejs) – Rebuilds a reference image object as a token-efficient, procedural, quality-gated TypeScript Three.js model ready for animation.

* [Agent Manager v0.33.0](https://github.com/YoanWai/agent-manager) – Terminal UI that runs multiple AI coding agents in separate tmux sessions with live status, project grouping, quick prompting, and file-diff review.

* [Nudgebee v1.6.0](https://github.com/nudgebee/nudgebee) – SRE copilot that watches Kubernetes and cloud accounts, ranks findings, and guides investigation, remediation, and cost optimization with chat and runbook automation.

* [Surf v2.16.0](https://github.com/nicobailon/surf-cli) – CLI for AI agents to control Chrome with zero-config, agent-agnostic browser automation.

* [NTM - Named Tmux Manager v1.30.0](https://github.com/Dicklesworthstone/ntm) – Tmux-based session manager that spawns, coordinates, and monitors multiple AI coding agents in tiled panes.

* [OKF v2.2.0](https://github.com/serradura/okf) – Durable, structured knowledge bundles for AI agents, stored as versioned Markdown files and consumed through a Ruby-based library with CLI, Graph, and MCP interfaces.

* [Dive v1.26.0](https://github.com/deepnoodle-ai/dive) – Foundational Go library for building AI agents and LLM-powered applications with multi-provider access, tool calling, and a robust agent loop.

* [Agent Swarm v1.135.0](https://github.com/desplega-ai/agent-swarm) – Multi-agent orchestration platform that runs lead and worker AI coding agents in Docker, coordinating tasks, integrations, and persistent memory.

* [Ouijit v1.7.0](https://github.com/ouijit/ouijit) – Git worktree-based task and terminal session manager integrating agent CLIs, lifecycle hooks, and VM sandboxing.

* [Forge v0.19.0](https://github.com/initializ/forge) – Secure, portable AI agent runtime that compiles Agent Skills from SKILL.md into hardened agents with outbound-only networking, encrypted secrets, and audit logging.

* [Kandev v0.91.0](https://github.com/kdlbs/kandev) – Kanban-based development environment that orchestrates multiple AI agents, reviews changes, and manages parallel tasks.

* [dsh-router-standard v0.3.0](https://github.com/yjh051108/dsh-router-standard) – Task-aware reasoning-mode router for DeepSeek Harness that quantizes persona behavior into three stable bands and switches between standard and spec modes per session.

* [Warren v0.18.0](https://github.com/jayminwest/warren) – Self-hostable control plane that spawns ephemeral, sandboxed coding agents on GitHub repos, streams events, and returns a branch for PRs.

* [VaneHub AI v1.0.0](https://github.com/cdavid817/vanehub-ai) – Desktop-first workspace that unifies AI coding agent CLIs and native HTTP agents in one React interface with clear runtime boundaries.

* [Sortie v1.21.0](https://github.com/sortie-ai/sortie) – Agent orchestrator that turns tracker issues into isolated, parallel autonomous agent sessions with persistent state, retries, CI feedback, and cost tracking.

* [AgentHound 1.1.0](https://github.com/adithyan-ak/AgentHound) – Offensive security framework for AI agent infrastructure that runs recon, credential and model inventory, exploitation, and builds a Neo4j graph to analyze attack paths.

* [ChatLab v0.36.2](https://github.com/ChatLab/ChatLab) – Private, local desktop app for analyzing and exploring chat histories using SQL and AI agents.

* [Stop That Shit 0.1.0](https://github.com/lennney/stop-that-shit) – Multi-platform hook and skill guard for AI coding agents, blocking unrequested hashes and checksums and preventing task-scope creep.

* [MiMoCode v0.1.13](https://github.com/XiaomiMiMo/MiMo-Code) – Terminal-native AI coding assistant with cross-session persistent memory, task tracking, multi-agent orchestration, and integrated Git and command execution.

* [J-Space Cognition Suite V3.6 v3.6.1](https://github.com/Tiger3807861189/J-Space-Cognition-Suite-V3.6) – Model-agnostic inference-time cognitive control suite for deep reasoning, long-horizon work, tool use, verification, and recovery via a managed workspace protocol.

* [ClawMem v0.37.0](https://github.com/yoloshii/ClawMem) – On-device context engine providing retrieval-augmented memory, hooks, and an MCP server for Claude Code, OpenClaw, and Hermes agents.

* [DAC v0.13.0](https://github.com/bruin-data/dac) – Dashboard-as-code for defining, validating, and serving dashboards from YAML and TSX with a semantic layer and AI agent.

* [Rampart v1.8.0](https://github.com/peg/rampart) – Security layer enforcing YAML policies for commands, file access, and network requests from AI coding agents.

* [DSH Desktop v0.3.0](https://github.com/dataelement/dsh-desktop) – Local-first, cross-platform desktop shell that packages the DeepSeek Harness web experience with managed launch, local networking, and persistent profiles, plugins, and sessions.

* [AgentOps v3.6.0](https://github.com/boshu2/agentops) – DevOps-style toolbox that preserves agent context, orchestrates workflows, and composes lasting knowledge between sessions.

* [numbat v0.2.0](https://github.com/perplexityai/numbat) – Endpoint visibility for AI agent activity with local detection, optional pre-action blocking, and forensic reconstruction from session artifacts and logs.

* [Smithers v0.35.0](https://github.com/smithersai/smithers) – Durable AI orchestration framework for defining workflows as React components.

* [agentjail v1.6.0](https://github.com/LuD1161/agentjail) – Policy guardrails for coding agents that locally check every tool call against OPA rules before execution.

* [Pad v0.14.0](https://github.com/PerpetualSoftware/pad) – Project management tool combining local-first CLI, web UI, and AI agent integration backed by SQLite.

* [IRIS v1.8.0](https://github.com/IRISX-AI/IRIS-AI) – Voice-first desktop AI assistant that executes real system actions using Gemini Live WebRTC audio, biometric security, and autonomous orchestration.

* [Axern v0.5.0](https://github.com/cofy-x/axern) – Axern provides an agent sandbox and durable service platform that isolates untrusted code and runs trusted long-lived workloads with a shared resource and lifecycle model.

* [Sigbound v2.4.0](https://github.com/surya-koritala/sigbound) – Sigbound runs multiple AI coding agents in parallel on one git repository, merging results only after verified builds and tests succeed.

* [managed-agents v0.3.0](https://github.com/sandbaseai/sandbase-harness) – Local-first runtime for AI agents with persistent sessions, sandboxed tools, SQLite-backed storage, audit/replay, and a built-in console UI.

* [SOLO v1.1.0](https://github.com/solo-agent/solo) – Local-first workspace that coordinates humans and AI coding agents with channels, threaded conversations, task boards, and channel-scoped teams.

* [clawk v0.4.0](https://github.com/clawkwork/clawk) – Disposable Linux VMs for autonomous coding agents, providing an allow-listed network sandbox and forwarded SSH access while keeping host files and secrets isolated.

* [claude-auto-retry v0.7.0](https://github.com/cheapestinference/claude-auto-retry) – Automatically retries Claude Code sessions after rate-limit or overload messages by waiting for printed reset times with safety margins and sending continue via tmux.

* [Agent Message Queue (AMQ) v0.73.0](https://github.com/avivsinai/agent-message-queue) – Local file-based interoperability bus that coordinates agent conversations, threads, routing, and delivery state with crash-safe maildir-style messages.

* [Nimbalyst v0.75.3](https://github.com/nimbalyst/nimbalyst) – Visual workspace for building and collaborating with coding agents across files, sessions, and tasks.

* [Agenvoy v0.32.15](https://github.com/agenvoy/Agenvoy) – Go-based AI agent framework with self-improving error memory, multi-provider LLM routing, script/HTTP tool extensions, and OS-native sandboxing.

* [skillreaper v0.7.0](https://github.com/thousandflowers/skillreaper) – Prunes unused AI-agent skills, MCP servers, subagents, hooks, and always-loaded prose from transcript evidence to reduce context bloat.

* [LiveAgent v1.2.7](https://github.com/Stack-Cairn/LiveAgent) – Local-first AI agent desktop client combining multi-model chat with local tool execution, MCP tool bridging, and a remote gateway for access and collaboration.

* [META\_KIM v3.0.6](https://github.com/KimYx0207/Meta_Kim) – Governance layer that coordinates AI coding assistants via an 8-stage workflow, contracts, gates, and memory to improve execution and learning.

* [Sandcastle v0.47.0](https://github.com/gizmax/Sandcastle) – Workflow orchestrator for AI agent pipelines defined in YAML, with multi-provider routing, integrations, and dashboard.

* [xlflow v0.31.0](https://github.com/harumiWeb/xlflow) – AI-agent-ready framework that turns Excel VBA workbooks and add-ins into a CLI-driven workflow for exporting, editing, linting, testing, tracing, and executing code.

* [Marinara Engine v2.4.4](https://github.com/Pasta-Devs/Marinara-Engine) – Local AI-powered chat, roleplay, and game engine offering plug-and-play setup.

* [ax v0.43.0](https://github.com/Necmttn/ax) – Local agent-experience loop that learns from AI coding agent sessions, records retro experiments in a typed graph, and proposes repo-specific fixes for review.

* [TanStack AI @tanstack/ai-lovable...](https://github.com/TanStack/ai) – Type-safe SDK for building provider-agnostic, multimodal AI applications with chat, streaming, agent loops, and isomorphic tools.

* [cass-memory v0.2.14](https://github.com/Dicklesworthstone/cass_memory_system) – Provides persistent, cross-agent procedural memory by converting scattered session logs into structured, confidence-tracked rules.

* [MemoraX Code v0.1.8](https://github.com/memorax-ai/memorax-code) – Memory plugin that adds cross-session repository knowledge, reusable lessons, and workflow preferences to multiple AI coding agents.

* [swarm-orchestrator v13.1.1](https://github.com/moonrunnerkc/swarm-orchestrator) – An agentic code runner that plans changes, records tool evidence through hash-chained logs, runs and retries gates with a numeric ratchet, and exports verifiable bundles.

* [AI Workdeck v0.25.0](https://github.com/zeweihan/aiworkdeck) – AI-native workspace for legal and document-heavy workflows, combining file management, agents, plugins, WPS editing, OCR, and evidence-chain workflows.

* [e2a v1.7.12](https://github.com/tokencanopy/e2a) – Authenticated email gateway for AI agents that verifies SPF, DKIM, and DMARC, then delivers evidence via webhook, WebSocket, REST, or MCP with optional human approval.

* [Claude Sidebar 1.9.5](https://github.com/derek-larson14/obsidian-claude-sidebar) – Embedded Obsidian sidebar terminal to run Claude Code with multiple tabs, folder-aware context, and YOLO mode.

* [MyAgents v0.4.11](https://github.com/hAcKlyc/MyAgents) – Desktop AI agent workspace and task center with multi-provider runtimes, IM/chat integrations, and CLI/SDK tooling.

* [OpenSwarm v0.21.1](https://github.com/unohee/OpenSwarm) – Autonomous orchestrator that runs Claude Code agents to process Linear issues, produce code changes, report to Discord, and store memory in LanceDB embeddings.

* [AgentReach v0.4.0](https://github.com/bojieli/agentreach) – AgentReach is a tool that proxies a coding agent’s shell and file operations over a single SSH connection so only the target is touched.

* [mediagen v0.3.0](https://github.com/Cripacx/mediagen) – Image and video generation skill that unifies Google Gemini, OpenAI, and Kie AI behind a single CLI and MCP server, with optional AI-content marking.

* [anywhere-agents v0.7.18](https://github.com/yzhao062/anywhere-agents) – Opinionated portable AI-agent configuration providing synced defaults, review workflows, and safety hooks across projects and sessions.

* [Claudexor v3.8.1](https://github.com/razzant/claudexor) – Local-first control plane that runs multiple AI coding agents via a typed interface, adding quota-aware routing, best-of-N races, and auditable execution evidence.

* [Daintree v0.33.0](https://github.com/daintreehq/daintree) – Daintree is a macro-orchestration app for running many AI coding agent terminals in parallel across git worktrees with context injection and workflow automation.

* [Agent Kanban v1.15.1](https://github.com/saltbo/agent-kanban) – Agent-first task board that assigns cryptographically identified AI coding agents to create, organize, and execute tasks with real-time team coordination.

* [Agentfiles 0.8.1](https://github.com/Railly/agentfiles) – Obsidian plugin to browse, create, and manage agent skills across 13+ coding assistants.

* [Autoprompt v1.0.4](https://github.com/Spielewoy/autoprompt-skill) – Coding-agent skill that reduces failure rates on agentic coding tasks by adding autoprompting to supported providers.

* [OpenChoreo v1.1.6](https://github.com/openchoreo/openchoreo) – Internal developer platform providing abstractions and orchestration for Kubernetes-based platform engineering teams.

* [ccmux v1.3.2](https://github.com/epilande/ccmux) – Daemon and TUI for tracking multiple AI coding agent sessions in tmux, showing live states and enabling fast switching and in-place actions.

* [Uni-CLI v1.2.1](https://github.com/olo-dot-io/Uni-CLI) – Command-grade access layer exposing websites, desktop apps, CLIs, and system capabilities as searchable, self-repairing commands for AI agents.

* [The Pair v2.7.1](https://github.com/timwuhaotian/the-pair) – Automated pair-programming app where two AI agents cross-check code and collaborate with real-time monitoring.

* [skill-up v0.9.1](https://github.com/alibaba/skill-up) – Evaluation and evolution tool for Agent Skills that runs declarative YAML cases across engines and iteratively repairs or expands eval suites based on failures.

* [REDCELL v0.2.0](https://github.com/martian56/redcell) – REDCELL orchestrates LLM agents to execute real penetration testing in a Kali container and generate structured reports with live operator control.

* [OCX v2.0.15](https://github.com/kdcokenny/ocx) – OpenCode extension manager providing portable, profile-based configs, isolated registries, and SHA-256-verified, version-pinned components.

* [nodeterm v0.3.2](https://github.com/eneskirca/nodeterm) – Node-based terminal and AI session manager that maps persistent tmux sessions onto a draggable infinite canvas with optional Trello-style project boards.

* [GMGN Agent Skills v1.5.7](https://github.com/GMGNAI/gmgn-skills) – Multi-chain AI-agent skills for real-time on-chain token analytics and natural-language execution of Solana, BSC, and Base trades with advanced order and wallet management.

* [OpenBot v0.0.1](https://github.com/CopilotKit/openbot) – An agent platform that runs inside your infrastructure, giving AI coworkers isolated browser computers and gated, audited tool actions via a single governance gateway.

* [CloudTerm v1.4.0](https://github.com/BradPerbs/cloudterm) – CloudTerm is an Electron-based terminal workspace that combines SSH, SFTP, Telnet, and Windows RDP with file transfers, port forwarding, and an AI session assistant.

* [Agent-Safe Pipeline v0.1.2](https://github.com/decionis/agent-safe-pipeline) – Reference architecture for safely executing AI-agent proposed actions via an independent authorization boundary with intent capture, policy decision, human verification, and a gated single-use executor.

* [Human Review v0.6.1](https://github.com/petergyang/human-review) – A visual editor for HTML and Markdown that lets users make direct changes and anchored comments, then sends all feedback to an AI agent in one batch.

* [Mirage v0.0.5](https://github.com/strukto-ai/mirage) – Unified virtual filesystem mounting diverse services and data sources under a single Unix-like tree for AI agents.

* [hypha v1.0.1](https://github.com/CodeSoul-co/Hypha) – TypeScript framework combining an agent core with a production harness to run governed, durable domain agents with checkpoints, recovery, and cache reuse.

* [Webcmd webcmd-v0.7.0](https://github.com/agentrhq/webcmd) – Self-learning browser infra that captures websites’ navigational context and compiles it into deterministic commands to make agent automation faster, cheaper, and more reliable.

* [Citadel v1.3.5](https://github.com/SethGammon/Citadel) – Operating layer that turns one-off coding-agent chats into repeatable workflows with durable memory, intent routing, safety hooks, telemetry, and coordinated agents.

* [GetVect v0.1.8](https://github.com/craigjmidwinter/getvect) – Local-first raster-to-vector desktop app that traces PNG, JPEG and BMP into SVG, EPS, DXF, PDF or PNG with offline processing and configurable quality controls.

* [AI Maestro v0.36.43](https://github.com/23blocks-OS/ai-maestro) – Agent orchestrator providing persistent memory, agent-to-agent messaging, and a dashboard to manage agents across multiple machines.

* [reminal v3.1.8](https://github.com/harshalgajjar/Reminal) – Reminal streams and mirrors a machine’s real windows and full terminal into any browser with end-to-end encryption and no open ports.

* [AtomicMemory cli-v0.2.1](https://github.com/atomicstrata/atomicmemory) – Inspectable, portable semantic memory layer for agents and applications, supporting portable capture, groundings, and correction-aware knowledge across SDK, CLI, MCP, adapters, and plugins.

* [Grok UI v0.12.1](https://github.com/joeynyc/Grok-UI) – Local-first dashboard for Grok Build that streams live runtime status, manages sessions via ACP, and shows Git changes and session history in-browser.

* [yorishiro v0.7.4](https://github.com/sktkkoo/Yorishiro) – Presence-focused macOS terminal that lets an AI inhabit and reshapes its scene and UI in real time through editable, self-modifiable packs.

* [WhatIfIBought v2.0.4](https://github.com/mamawai/wtfibought) – Live arena where LLM trading agents trade real market data with virtual money while exposing every thought, tool call, and post-trade review.

* [Superset desktop-canary](https://github.com/superset-sh/superset) – Terminal for running and managing 10+ parallel CLI coding agents locally.

* [Kane CLI 0.8.6](https://github.com/LambdaTest/kane-cli) – Validation layer for AI coding agents and natural-language browser automation that launches Chrome and returns structured run results.

* [Juggler v0.5.8](https://github.com/juggler-ai/juggler) – A visual AI coding agent with a local, shared tree-based session, inspectable tool calls, and plugin-driven UI and orchestration.

* [San v1.22.5](https://github.com/genai-io/san) – Terminal-native unified runtime for specialized AI agents built on pluggable LLMs, search backends, personas, and skills.

* [Mister Morph v0.3.14](https://github.com/quailyquaily/mistermorph) – Unified agent CLI and reusable Go agent core.

* [Maestro-Flow v0.5.82](https://github.com/catlog22/maestro-flow) – Workflow orchestration framework that routes intent to adaptive multi-agent command chains, runs parallel agents, and provides a real-time dashboard and evolving knowledge graph.

* [7 (Gnuboard7) 7.0.9](https://github.com/gnuboard/g7) – Gnuboard7 is a modern, extensible CMS platform built with Laravel and React, featuring JSON-driven React UI and role-based permissions.

* [AI Coding Project Boilerplate v1.27.5](https://github.com/shinpr/ai-coding-project-boilerplate) – Production-ready TypeScript boilerplate optimized for Claude Code, using specialized sub-agents to prevent context exhaustion.

* [Agent SDK for Go v0.3.5](https://github.com/agenticenv/agent-sdk-go) – Temporal or Restate-backed Go SDK for running AI agents durably, with pluggable components for state, tool calls, and execution loops across restarts.

* [caam - Coding Agent Account Manager v0.1.17](https://github.com/Dicklesworthstone/coding_agent_account_manager) – Sub-100ms auth switching for AI coding CLIs to instantly swap subscription accounts when usage limits are reached.

* [GoAI v0.9.7](https://github.com/zendev-sh/goai) – Go SDK providing a unified API for 25+ AI providers with streaming, structured output, and MCP support.

* [IM.codes android-dev-v1.1-117...](https://github.com/im4codes/imcodes) – IM.codes provides a shared memory layer and managed MCP tool surface for coding agents, including supervised execution and cross-agent audit across AI providers.

* [ClawRouter v0.12.246](https://github.com/BlockRunAI/ClawRouter) – Smart LLM router that locally routes requests to the cheapest capable model across 30+ providers, reducing inference costs.

* [AgentDock v0.7.7](https://github.com/uvwt/agentdock) – Secure agent runtime that provides unified, permissioned access to files, commands, Git, Skills, MCP, and browser automation across local, remote, and container environments.

* [DeepSeek Harness dsh-v0.1.1-rc.2](https://github.com/deepseek-ai/deepseek-harness) – DeepSeek Harness is an agent harness built on Cordis, using an architecture where everything is a plugin.

* [Hivemind v0.7.148](https://github.com/activeloopai/hivemind) – Persistent cloud-backed shared memory for Claude Code, OpenClaw, and Codex agents.

* [Autopus-ADK v0.50.108](https://github.com/Insajin/autopus-adk) – Harness that orchestrates multiple AI coding agents into engineering workflows with planning, testing, code review, and security audits.

* [Incognide v0.2.28](https://github.com/NPC-Worldwide/incognide) – Unified tileable workspace combining chat, code, documents, web browsing, and media with context-aware automations.

* [Codra v0.9.5](https://github.com/devarshishimpi/codra) – Self-hosted AI code review system for GitHub pull requests, running on Cloudflare Workers with queue-backed processing and a repository-aware dashboard.

* [KiwiFS v0.19.62](https://github.com/kiwifs/kiwifs) – Knowledge server exposing Markdown files with Git versioning, full-text and vector search, and an Obsidian-like web UI.

* [chartr v0.2.4](https://github.com/rengwu/chartr) – Agent multiplexer that visualizes work as a star-map and coordinates agent sessions with contextual tickets and blockers.

* [Poracode v1.6.5](https://github.com/SDSLeon/lightcode) – Poracode is a desktop app that runs multiple AI coding agents side-by-side with shared terminal, chat, flexible layouts, and built-in MCP-based orchestration.

* [Nylas CLI v3.1.30](https://github.com/nylas/cli) – Command-line tool for Nylas API authentication and managing email, calendar, contacts, webhooks, timezone utilities, and OTP extraction.

* [bkit - Vibecoding Kit v2.1.38](https://github.com/popup-studio-ai/bkit-claude-code) – Claude Code plugin providing PDCA-based development workflows, systematic context engineering, CTO-led multi-agent orchestration, and intelligent code assistance.

* [bkit v2.1.38](https://github.com/ww-w-ai/bkit-claude-code) – A Claude Code plugin that verifies AI-generated code against design specs and auto-repairs mismatches using multi-agent skills, quality gates, and persistent memory.

* [EverFern v1.1.6](https://github.com/Everfern-AI/Everfern) – Local desktop AI agent that plans and executes computer actions, including browser research, with sandboxed execution and persistent memory.

* [Emend v0.1.1](https://github.com/diyanbogdanov/emend) – Emend performs verified dependency migrations by applying version bumps in throwaway worktrees, rerunning checks, and proving the code still compiles and tests pass.

* [AI Team OS v1.11.3](https://github.com/CronusL-1141/AI-company) – Multi-agent operating system that runs autonomous AI teams to execute, coordinate, and self-improve software projects.

* [COMandA v0.0.219](https://github.com/kris-hansen/comanda) – Command-line tool to define, version-control, and run YAML-based LLM pipelines locally or as servers.

* [Ship Studio v0.18.6](https://github.com/ship-studio/ship-studio) – Local-first desktop app that combines an AI agent terminal, live preview, visual editing, Git workflows, and one-click Vercel deploys.

* [ZeroID v1.8.8](https://github.com/highflame-ai/zeroid) – Identity infrastructure that issues verifiable agent credentials, enforces delegated on-behalf-of chains, and enables real-time revocation.

* [PR Cockpit nightly](https://github.com/taovc/pr-cockpit) – Local AI PR workbench that uses Claude or Codex to review PR queues in isolated worktrees, with human-gated GitHub inline comments.

* [Rove v0.8.196](https://github.com/Sma1lboy/rove) – Terminal-native workspace that runs multiple coding tasks in parallel using isolated git worktrees and persistent agent or shell sessions.

* [Wallfacer v0.0.24](https://github.com/changkun/wallfacer) – Self-operating engineering platform orchestrating autonomous coding agents across a task board with isolated sandboxes, live logs, diffs, cost tracking, and full oversight.

* [Orkas v2026.8.25](https://github.com/Orkas-AI/Orkas) – Local-first desktop app for commanding a team of AI agents from one chat, using a commander agent to coordinate specialized sub-agents with your own LLM keys.

* [agent-inspect @agent-inspect/vites...](https://github.com/rajudandigam/agent-inspect) – Local evidence debugger for TypeScript AI agents that turns runs into execution trees for debugging, regression checks, and share-checked offline artifacts.

* [Machinist factory-legacy-final](https://github.com/owainlewis/machinist) – Runtime and control plane that runs a supervised local team of coding agents to turn GitHub issues into reviewed pull requests.

* [20x v0.0.136](https://github.com/peakflo/20x) – Desktop app turning task lists into AI-powered agents that triage, execute, and manage tasks locally.

* [Dexto @dexto/tools-filesys...](https://github.com/truffle-ai/dexto) – Universal intelligence layer that orchestrates LLMs, tools, and data to build stateful, multimodal agentic applications.

* [HackerAI desktop-v0.0.59](https://github.com/hackerai-tech/hackerai) – AI-powered penetration testing assistant providing automated reconnaissance, exploit guidance, and reporting.

* [Genie v5.260816.1](https://github.com/automagik-dev/genie) – CLI that interviews you, builds a plan, runs parallel agents in isolated worktrees, and outputs reviewed pull requests.

* [Thunderbolt v0.1.123-nightly.202...](https://github.com/thunderbird/thunderbolt) – Cross-platform AI client for on-prem and local models with enterprise features and self-hosting support.

* [OpenCove v0.2.0-nightly.20260...](https://github.com/DeadWaveWave/opencove) – Infinite 2D canvas workspace that organizes AI agents, terminals, tasks, and notes.

* [\_mom\_ v0.53.0-alpha](https://github.com/momhq/mom) – Local-first persistent memory layer for AI coding agents, storing decisions and tool transcripts for recall across sessions and tools.

* [AEGIS aegis-v0.13.0-alpha](https://github.com/antropos17/Aegis) – Local-first EDR-style monitor tracking AI agent processes, file and network activity with per-agent risk scoring.

* [open-sheet @open-sheet/core@0.1...](https://github.com/lianghsun/open-sheet) – A spreadsheet framework that lets agents write React-based models, handling cell addressing, formula references, recalculation, and exporting live Excel workbooks.

* [KaaS v0.1.0-rc.2](https://github.com/bybit-exchange/kaas) – KaaS compiles scattered notes, documents, and transcripts into a searchable Markdown wiki using a multi-phase LLM pipeline with MCP access, no embeddings.

* [Myrlin's Workbook v1.3.0-alpha.30](https://github.com/therealarthur/myrlin-workbook) – Workspace manager for Claude Code sessions with multi-pane embedded terminals, project discovery, kanban board, cost tracking, and session templates.