The index · everything in the almanac
Search
Look up a team, a repository, a gathering or a paper, by keyword or by meaning.
No matches for “”
The index · everything in the almanac
Look up a team, a repository, a gathering or a paper, by keyword or by meaning.
No matches for “”
No matches for “”
Layer
Startupkeyword match
The de-facto standard toolkit for building LLM applications and agents. · San Francisco, CA
LangChain makes the LangChain framework and LangGraph, the low-level orchestration standard for stateful, controllable agents, plus the LangSmith platform for tracing, evaluating and monitoring them in production.
Startupkeyword match
Conversational AI agents for customer experience. · San Francisco, CA
Sierra, co-founded by Bret Taylor and Clay Bavor, builds production conversational agents that resolve real customer issues with guardrails, evaluation and human escalation — a template for deploying vertical agents at the enterprise edge.
Open sourcekeyword match
LangChain/langgraph · Python
LangGraph models agents as graphs of nodes with checkpointed state, cycles, human-in-the-loop interrupts and durable execution. It has become the reference architecture for production agent orchestration.
Open sourcekeyword match
Anthropic/claude-agent-sdk-python · Python
Anthropic's Agent SDK packages the harness that powers Claude Code — the tool-use loop, subagents, context management, MCP support and hooks — so developers can build the same class of coding and task agents on top of Claude models.
Open sourcekeyword match
Mastra/mastra · TypeScript
Mastra is a full TS-native agent framework — workflows, tool calling, RAG, evals and a dev playground — that has become the default choice for JavaScript teams building agent backends.
Open sourcekeyword match
OpenClaw Foundation/openclaw · TypeScript
OpenClaw (born as Clawdbot in late 2025) is a self-hosted personal agent that connects your chats, email, browser and terminal into one always-on assistant. Its explosive growth — hundreds of thousands of stars within months — made 'personal agent harness' a mainstream category.
Open sourcekeyword match
DeepSeek/deepseek-harness · TypeScript
DeepSeek Harness (dsh) is DeepSeek's open-source agent harness, built on an everything-is-a-plugin architecture powered by the Cordis framework. It runs as a local Web UI via npx @deepseek-ai/dsh web, with desktop builds for macOS and Windows. It is a developer preview: every release so far (0.2.x) is a pre-release and the README warns of breaking changes.
Startupkeyword match
Multi-agent orchestration, from open-source framework to enterprise platform. · Coral Gables, FL
CrewAI popularized role-based multi-agent teams ('crews') with a simple mental model and ships an enterprise control plane for deploying, monitoring and governing them.
Startupkeyword match
The data framework for connecting agents to your data. · San Francisco, CA
LlamaIndex provides ingestion, indexing and retrieval primitives that give agents grounded context over enterprise data — a foundational piece of the retrieval side of the harness stack.
Startupkeyword match
Stateful agents with memory, from the team behind MemGPT. · Berkeley, CA
Letta (formerly MemGPT, out of UC Berkeley) builds an agent server with built-in memory, context management and state — letting agents persist, learn and be edited like files rather than living in a chat window.
Startupkeyword match
Durable execution for long-running agent workflows. · Seattle, WA
Temporal's durable execution engine is widely adopted to make agent pipelines reliable: state survives crashes, retries are automatic, and workflows run for minutes or months. A key piece of serious harness infrastructure.
Startupkeyword match
Durable functions and workflows for AI agents. · San Francisco, CA
Inngest runs agent logic as durable steps — queueing, retries, concurrency control and human-in-the-loop waits — without operating your own workflow infrastructure.
Startupkeyword match
Design, deploy and manage AI assistants for companies. · Paris, France
Dust (Paris) is an enterprise platform for building assistants and agents connected to company knowledge and tools — strong on developer experience and European enterprise adoption.
Startupkeyword match
AI employees that run your workflows. · San Francisco, CA
Lindy lets anyone build 'AI employees' — email triage, meeting prep, support — with an event-driven agent runtime and a large template library.
Startupkeyword match
Build an AI workforce that runs itself. · Sydney, Australia
Relevance AI (Sydney/SF) provides a platform for building, orchestrating and monitoring teams of work agents — 'tools' and 'teams' primitives for non-developers.
Startupkeyword match
A natural-language IDE for building agents. · London, UK
Wordware (YC) lets domain experts and engineers co-build agents in a document-like natural-language programming environment — 'Notion for prompt-to-agent development'.
Startupkeyword match
Orchestration for swarms of enterprise agents. · New York, NY
Emergence AI builds an orchestration layer where specialized agents (web, code, data) are dynamically composed into workflows — 'an operating system for agent swarms'.
Startupkeyword match
The TypeScript-first company behind the Mastra agent framework. · United States
Mastra builds the eponymous open-source TypeScript framework — workflows, tools, RAG, evals and a local dev playground — and is becoming the default 'agent backend' choice for JS/TS teams.
Startupkeyword match
The programming language for agentic software. · United States
Agno (formerly Phidata) builds a high-performance open-source agent runtime — memory, knowledge, storage and monitoring built in — aiming to be the substrate beneath agentic applications rather than just a framework.
Startupkeyword match
Enterprise AI platforms around the Haystack framework. · Berlin, Germany
deepset (Berlin) builds the Haystack framework and an enterprise platform for industrial-grade RAG and agent pipelines — a European anchor of the NLP/agent engineering space since 2018.
Startupkeyword match
Type-safe Python infrastructure — now including agents.
Pydantic builds the data-validation library underlying most of the Python AI ecosystem, plus Logfire (observability) and Pydantic AI (typed agent framework) — the type-safety layer of the harness.
Startupkeyword match
Open-source TypeScript agent engineering platform. · United States
VoltAgent ships a batteries-included open-source TS agent framework — agents, workflows, observability UI and evals — targeted at product teams that want the JS ecosystem's answer to LangGraph.
Startupkeyword match
The durable runtime for agents and backends. · Zurich, Switzerland
Restate is a lightweight durable-execution runtime purpose-built for AI agents and workflows — journaling every step so invocations survive crashes, retries are deterministic, and long-running agent tasks run reliably. Raised a $20M Series A in September 2026 to 'define the infrastructure layer for AI agents'.
Startupkeyword match
Trust and infrastructure for open-source ecosystems — now Mecatl. · Seattle, WA
Stacklok, founded by Kubernetes/Sigstore creator Craig McLuckie, builds trust tooling for open source. In 2026 it shipped Mecatl — an open-source, cloud-native agent harness that architecturally separates the agent loop from code execution, built to run identically on a laptop or Kubernetes.
Startupkeyword match
Boutique lab crafting tools to harness AI — Pi and Lefos. · Remote
Earendil Works 'crafts tools to harness AI': Pi, the minimal, steerable agent toolkit (unified LLM API, agent loop, TUI) that became a favorite base for custom harnesses, and Lefos. Built in the open, shaped by the humans who use them.
Startupkeyword match
Raven — the harness of harnesses, built for recursive self-improvement.
EverMind AI builds Raven, an open-source, trusted, persistent and self-evolving multi-agent ecosystem designed for recursive self-improvement (RSI), alongside memory-sparse-attention research for scaling end-to-end memory to 100M tokens.
Startupkeyword match
A multiplayer agent harness for work.
QM gives every person and shared room a durable, isolated agent workspace while keeping administration, permissions and background work centralized — docs cover models and harnesses, sandboxes, apps and skills, and Slack/web surfaces. Its open-source runtime (open-multi-agent) ships durable approvals and verifiable run records.
Startupkeyword match
An engineering 'superharness' that runs coding agents as one governed team.
Palgu describes itself as an engineering superharness that takes a task from intent to a proven, governed code change. It runs coding agents such as Claude Code, Codex, Cursor and aider as one team and wraps them in eight layers: intent, architecture, knowledge, coordination, evidence, recovery, governance and learning. Its products are the palgu CLI and the Palgu Studio desktop app for macOS and Windows, plus open-source Maggy (a single-developer harness), the Mixture of Models mod for Claude Code and other agents, and AWG, a shared work graph for coding agents. Features include a code graph with blast-radius mapping, a five-seat review council of models other than the author, an audit ledger, BYOK and a self-hosted gateway.
Startupkeyword match
p0: a desktop app that orchestrates coding agents to bring software development in-house. · San Francisco, CA
BePurple, Inc. makes p0, a desktop app for macOS and Windows that turns a feature description into a spec and mockups and then orchestrates coding agents to build it. It runs on Claude Code or Codex, and on any endpoint compatible with the Anthropic Messages API, such as Kimi, DeepSeek, GLM or OpenRouter. It splits work into phases, runs agents in parallel with file ownership in isolated Git worktrees, and uses QA agents to check acceptance criteria before a PR opens. It also generates codebase Standards and has a Custom Agents Studio. Code generation runs locally, and model usage is billed by the user's provider.
Startupkeyword match
Stably AI's open-source environment for running fleets of coding agents in parallel worktrees. · San Francisco, CA
Orca is an MIT-licensed agent development environment from Stably AI, a Y Combinator-backed San Francisco company. It runs coding agents such as Claude Code, Codex, OpenCode, Pi, Gemini and Cursor CLI side by side, with about 27 agents supported. Each task gets its own isolated git worktree, so one prompt can fan out to several agents and the best result can be merged. Other features are GPU-rendered terminal splits, a per-worktree Chromium 'Design Mode', GitHub and Linear integration, SSH remote runtimes, an Orca CLI that agents can drive, and iOS/Android companion apps.
Startupkeyword match
Open-source orchestration for running a company of AI agents: org charts, goals, budgets and governance.
Paperclip, from Paperclip Labs, Inc., is an MIT-licensed, self-hostable Node.js server and React UI that orchestrates teams of AI agents around business goals. It pitches itself as 'If OpenClaw is an employee, Paperclip is the company'. Users bring their own agents and harnesses (OpenClaw, Codex, Claude, Cursor and others) and choose them per agent. Paperclip adds org charts and reporting lines, goal-linked tasks, scheduled heartbeats and routines, cost budgets with threshold pauses, approval stages, shared skills, connector access control and multi-org isolation.
Startupkeyword match
Agent collaboration platform that hires, schedules and reports on a team of AI agents.
LobeHub, formerly the LobeChat project, describes itself as an agent collaboration platform for work. It casts itself as a 'Chief Agent Operator' that runs agents around the clock: it hires and schedules them and reports on the work. Its focus is persistent agent teammates and group collaboration with shared context. The source is on GitHub (repo created 2023) under the LobeHub Community License. That license is Apache 2.0 with extra conditions, and distributing derivative works needs a commercial licence. Users can self-host with Docker or Vercel, or use the hosted LobeHub Cloud at app.lobehub.com.
Startupkeyword match
Open-source cowork desktop app that runs a local multi-agent workforce.
Eigent is an open-source desktop 'cowork' application for building and running an AI workforce that automates multi-step workflows. Built on the CAMEL-AI open-source framework, it offers a single-agent harness for focused tasks and a multi-agent workforce mode for parallel execution, with MCP and skill integration, built-in browser and terminal toolkits, automations and support for any model, including local ones. It can be self-hosted for free with your own keys; paid cloud plans add task credits, and enterprise features include SSO and access control. The site's legal entity is Eigent UK Ltd.
Startupkeyword match
Agentic IDE for running Claude Code, Codex and other CLI agents in parallel worktrees.
Paradev (formerly Superset; unrelated to Apache Superset) is a desktop workspace for running CLI coding agents such as Claude Code and Codex in parallel, each in its own git worktree with its own branch, terminals and per-workspace browser preview. It includes a diff viewer, PR review that sends selected lines back to an agent, a Design mode for UI changes, and shared Pages for team review. Users bring their own agent subscriptions and the app does not proxy model API calls. Source is public under the Elastic License 2.0 (source-available, not OSI open source); macOS is the primary platform. Its site announces an $11.5M seed.
Startupkeyword match
ART: open-source reinforcement learning trainer for multi-step agents.
OpenPipe maintains ART (Agent Reinforcement Trainer), an open-source (Apache-2.0) RL framework that lets LLM agents learn from experience by integrating GRPO into any Python application. Project news covers a LangGraph integration, MCP-RL for training models to use MCP server tools, AutoRL for zero-data training and RULER for LLM-judged rewards. The README also promotes W&B Training (Serverless RL), a managed service for running ART training and inference. Supported open models include Qwen, GPT-OSS and Llama.
Startupkeyword match
TrueForge: open-source agent harness runtime layer.
TrueFoundry, which sells an enterprise AI gateway and MCP gateway, maintains TrueForge, an open-source (MIT) agent harness that runs the agent execution loop: model calls, MCP tools (with header auth or OAuth), git-backed SKILL.md skills, sandboxed code execution (Daytona today), tool approvals, context management and session state. TrueForge exposes this through a chat UI, an HTTP API with a TypeScript SDK and an embeddable UI SDK, and adds context engineering such as subagents, deferred tool loading, Code Mode and compaction. It runs locally (SQLite) or hosted (Postgres + Redis).
Startupkeyword match
Open-source platform for agentic workflows and RAG pipelines, cloud or self-hosted.
Dify, developed by LangGenius, is an open-source LLM app development platform combining a visual workflow canvas, RAG pipelines, agent capabilities, model management across hundreds of proprietary and open models, a prompt IDE and observability integrations (Opik, Langfuse, Arize Phoenix). It is offered as Dify Cloud or self-hosted via Docker Compose. The source is under a modified Apache 2.0 license that requires a commercial license for operating multi-tenant services, among other conditions. Its GitHub repository has around 160k stars.
Startupkeyword match
Always-on software delivery agents built on shared codebase context. · Palo Alto, CA
Augment Code (Augment Computing Inc.) builds AI coding agents for professional engineering teams. Its Cosmos platform is pitched as an always-on software delivery system from spec to verification, with pre-built agent loops for code review, ticket-to-PR, vulnerability remediation and incident response. Work is triggered from GitHub, GitLab, Slack, Linear, Jira, schedules or webhooks, runs in managed cloud sandboxes or customer-managed machines, and carries permissions, audit trails, observability and human review points, with per-task model choice. Founded in 2022 by Igor Ostrovsky and Guy Gur-Ari.
Startupkeyword match
Agentic terminal and open infrastructure for cloud software factories.
Warp, founded in June 2020 by former Google principal engineer Zach Lloyd, began as a Rust-based terminal and now positions itself as open infrastructure for cloud software factories. Teams define a factory in code (factory.yaml plus agent files); agents triggered by issues, chat or schedules take work from triage to a mergeable PR, on Warp's cloud or self-hosted in a VPC, and can be steered from web, mobile, terminal or IDE. It supports any MCP-capable agent, including Claude Code, Codex and Cursor, and bring-your-own model or harness. In April 2026 Warp open-sourced its terminal and its Oz cloud-agent orchestration platform.
Startupkeyword match
A self-hosted worker that turns tracker tickets into pull requests.
DevIntern is a self-hosted daemon that picks up ready tickets from Jira, Linear, GitHub Issues and other trackers, routes each to the right repository, and runs the user's existing coding agent (Claude Code, Codex, Cursor or OpenCode) in isolated worktrees, with optional quality gates and auto-review, before opening a pull request. Review comments go back to the same worker as commits on the same branch. Interactive terminal use is free; unattended automation needs a paid license after a 14-day pilot. The source is public under FSL-1.1-ALv2, and the project works its own backlog in public.
Startupkeyword match
A self-hosted control layer for fleets of CLI coding agents.
AgentsMesh calls itself an 'AI Agent Workforce Platform' for running many coding agents across your own machines from one console. Self-hosted runners report their capacity and schedule agent pods, each with its own Git worktree sandbox, credentials and branch. Web, Electron desktop and iOS clients stream terminals in real time, and an Autopilot control agent sends each idle pod its next instruction, with iteration caps, decision history and human takeover. Supported agents include Claude Code, Codex CLI, Gemini CLI, Aider and OpenCode. The Go codebase is source-available under BSL-1.1, converting to GPL-2.0-or-later.
Startupkeyword match
Spec generation plus autonomous Ralph loops on top of Claude Code or Codex.
Wiggum CLI is a command-line orchestrator for autonomous coding loops, also called Ralph loops. It scans a codebase's stack, generates feature specs through AI-led interviews, and runs plan, implement, test, verify and PR cycles using Claude Code, Codex CLI or another CLI coding agent. An agent mode reads the GitHub backlog, prioritizes issues by label and dependency order, resumes partial branches, reviews diffs against the spec and auto-merges when checks pass. It can run headless in CI. It installs via npm, and the TypeScript source is licensed MIT with the Commons Clause.
Startupkeyword match
An infinite-canvas desktop workspace for running parallel coding agents.
AgentGrid is a desktop app for macOS, Windows and Linux that puts multiple coding agents (Claude Code, Codex, Cursor, OpenCode and others) on one infinite canvas alongside terminals, browsers and notes. A Coordinator pane takes natural-language instructions to create agents and teams, such as a builder and a QA reviewer, arrange the workspace and direct work across the canvas. Users bring their own AI subscriptions, and the canvas is shared so teams can follow every agent and review results together.
Startupkeyword match
Run a team of coding agents in parallel isolated workspaces, locally or in the cloud. · San Francisco, CA
Conductor creates isolated copies of a codebase, runs coding agents in each, and lets developers review, open PRs for and merge their work. It runs the first-party Claude Code, Codex, Cursor and OpenCode agents with the user's own subscriptions and keys. Conductor Cloud gives each agent a Firecracker microVM sandbox with repos and dependencies pre-installed, and workspaces are multiplayer. There are macOS and iOS apps and an API. Founded by Charlie Holtz and Jackson de Campos (YC S24), it raised a $22M Series A from Spark and Matrix in March 2026.
Startupkeyword match
An IDE and dashboard for orchestrating fleets of cloud coding agents. · San Francisco, CA
Capy says it was built from scratch for multi-agent orchestration. Developers assign feature, bug and refactor tasks to parallel coding agents, each on its own branch and sandboxed cloud VM, and the agents plan, build and open pull requests that are reviewed as normal. It ships desktop apps for macOS, Windows and Linux plus a web app, a Slack integration, automations, and end-to-end testing with native computer use, browser use and annotated videos. Capy (YC F24) was founded in 2024 in San Francisco by Nalin Semwal and Justin Sun.
Startupkeyword match
AI orchestration platform with spec-driven coding workflows and IDE agents.
Zencoder (For Good AI Inc.) offers Zenflow Code, pre-built workflows where agents draft specs, implement changes in isolated environments and run verification before review, with parallel execution, scheduled automation and cross-agent review; Zenflow Work, goal-driven automations over Jira, Slack, Notion and other tools; and IDE agents for VS Code and JetBrains. It describes multi-model workflows (plan with Claude, build with Gemini, review with Codex), multi-repo indexing, MCP support, BYOK and SSO.
Startupkeyword match
Autonomous enterprise software development with thousands of agents in parallel.
Blitzy builds an autonomous software development platform for large enterprises with sprawling legacy codebases. Per SiliconANGLE, it reverse-engineers existing environments into a dynamic knowledge graph of the codebase and uses an orchestration layer to run thousands of agents in parallel for days or weeks of inference, calling models from Google, Anthropic and OpenAI over 100,000 times in a single run. Founded in 2023 by Brian Elliott and Sid Pardeshi; it raised $200M at a $1.4B valuation led by Northzone in May 2026.
Startupkeyword match
Slate, a swarm-native coding agent for long-running engineering work. · San Francisco, CA
Random Labs (YC S24) builds Slate, a long-running autonomous coding agent. Slate V1, launched in March 2026, uses a 'Thread Weaving' design in which a central orchestration thread dispatches parallel worker threads through a TypeScript-based DSL; workers return compressed 'episodes' rather than full transcripts, and dynamic pruning manages context in large codebases. Different models can be assigned to orchestration and execution roles, and the CLI uses usage-based credits. Co-founded in 2024 by Kiran and Mihir Chintawar in San Francisco.
Startupkeyword match
General-purpose cloud agent with its own computer, acquired by Meta. · Singapore
Manus is a general-purpose AI agent launched in 2025 by Butterfly Effect, which was founded in China before moving to Singapore. Subscribers use it for research, coding and other long tasks, and its current site lists building slides, websites, games and video, Wide Research, a Slack integration and an API, across web, desktop and mobile apps. Meta announced its acquisition of Manus on December 29, 2025; terms were undisclosed (the WSJ reported over $2B), and Manus said it would keep operating its subscriptions.
Startupkeyword match
Agentic Work Graph: task management for AI agents, with atomic claims, parallel lanes and proof of completion.
AWG is a work graph for coding agents: intent, requirements, tasks, dependencies, decisions, validation, evidence and outcomes in one shared source of truth across sessions, repositories, tools and machines. Agents claim work atomically (leases with fencing tokens), take parallel lanes that do not collide, and close a task only when its completion contract is met with evidence, not by assertion. People see what is ready, blocked, running, proven or waiting for a decision in one ledger. It runs locally on SQLite with no account, self-hosted on PostgreSQL, or as AWG Cloud at awg.palgu.ai with a Palgu account. A Claude Code mod and an MCP server for Codex and other clients expose tools such as awg_next, awg_submit and awg_create.
Startupkeyword match
Messaging between coding agents: every agent gets an address, across people, vendors and tools.
clawalkietalkie gives each coding agent an address, such as cc: for Claude Code, cx: for Codex, cu: for Cursor, gm: for Gemini CLI and gh: for GitHub Actions, so agents can hand work to each other, including a teammate's agent at a team address, and report what they are working on to one status board. CI can message the agent that can fix a failing build and wait for the reply. Claude Code's mod starts a turn when a message arrives; other agents read their inbox through an MCP server. The clients are open source (MIT) and talk to a hosted relay.
Startupkeyword match
Nerve Center for Agentic Coding
Traycer is a free, open-source desktop app for Mac, Windows and Linux. It runs Claude Code, Codex, Cursor and OpenCode side by side, each in its own worktree. Agents share specs and tickets and can message each other to ask questions, request reviews and hand off work. Users bring their own agent subscriptions.
Startupkeyword match
Agentic Coding with Multiple Parallel Agents
Verdent AI is an agentic coding platform that runs work from a natural-language description and coordinates parallel agents. It ships as a Mac app and as VS Code and JetBrains extensions. The site positions it for building complete products, including login, data storage, payments and admin tools.
Startupkeyword match
Managed agent platform: pick the harness, model and tools; Ellipsis runs it
Ellipsis runs coding agents on its own infrastructure and handles VMs, permissions, budgets and logs. Agents are defined in YAML: trigger, harness and model, prompt, repos, hooks, compute and spend limits. Supported harnesses include Claude Code, Codex, OpenCode, Gemini CLI, GitHub Copilot and Cursor. Use cases are software factories, code review and background agents. The company is backed by Y Combinator.
Startupkeyword match
The free coding agent, built on the open Codebuff multi-agent framework
Codebuff is a YC-backed terminal coding agent, installed with `npm install -g codebuff`. Its open repo now ships as Freebuff, built on Codebuff, which the README calls the open multi-agent framework behind its orchestration, tools and SDK (@codebuff/sdk). Specialized agents find files, implement, review, research docs and test in a browser, and can run in parallel in isolated workspaces.
Startupkeyword match
Google's agentic development platform
Google Antigravity is an agentic development platform with four parts. The App is a command center for running multiple local agents in parallel. The CLI is terminal-first, with background subagents. The IDE is agentic, with an agent manager and artifacts. The Python SDK is for prototyping custom agents on the Antigravity harness.
Startupkeyword match
GitLab's platform for orchestrating AI agents and flows across the software lifecycle.
GitLab Duo Agent Platform provides specialized agents for planning, coding, security analysis and analytics that use GitLab context, plus a catalog for managing them. Flows chain agents into multi-step automations, such as diagnosing and fixing a broken CI/CD pipeline. Flows can be triggered by GitLab events, schedules or external tools, and include human approval checkpoints. It is offered to Premium and Ultimate customers on GitLab 18.8 or later.
Startupkeyword match
Build your own AI software factory.
Port is a platform for building, governing and operating an agentic software development lifecycle. It discovers agents, MCP connectors and skills across an organization so teams can enforce their standards. Agents handle tasks such as provisioning infrastructure, resolving tickets and fixing pipeline failures, under governance controls with humans kept in the loop.
Startupkeyword match
Command your own software factory.
Macroscope offers Murmur, in beta, which orchestrates cloud agents that write code and verify their work in isolated sandboxes. Its Code Review feature checks each change for correctness, security, tests and regressions before merge. Its Status feature shows what changed, why, and what needs attention across a team's repositories.
Startupkeyword match
AI Software Factory for Product Teams
Tempo turns customer feedback, issues and product requirements into reviewed pull requests. AI agents do the work while the team approves key decisions. It is billed as the IDE for the whole team.
Startupkeyword match
Browser platform and framework for AI web agents
Notte provides cloud browser sessions with anti-detection, residential proxies and CAPTCHA solving, plus browser agents that navigate sites from natural-language tasks. It can deploy browser automations as serverless functions with cron scheduling, and offers personas and secret vaults so credentials stay away from the LLM. It has an MCP server, a CLI and an agent skill for coding agents. The site says it is backed by Y Combinator (W25).
Startupkeyword match
AI Cloud in one line of code · San Francisco, CA
Langbase is a serverless platform for building, deploying and scaling AI agents, which are exposed as APIs (pipes) and can connect to 600+ LLMs, tools and MCP servers. It provides a managed memory/RAG service, and multi-step multi-agent workflows with retries, timeouts and step-level traces. It also offers Studio for versioning agents, plus tracing, cost tracking and evals. Its site also promotes Command Code, a coding agent.
Startupkeyword match
Discover, secure, govern, and optimize every AI agent.
Airia is an enterprise AI security and governance platform that discovers the models, agents, tools and MCP servers a company uses, including unapproved ones. It lets teams build and run agents across models and frameworks, and enforces agent actions at runtime by allowing, blocking, requiring approval, redacting context or logging before a tool call runs.
Startupkeyword match
The orchestration engine for teams who ship
Hatchet is a platform for orchestrating AI agents, background tasks and mission-critical workflows written as code in TypeScript, Python or Go. It handles retries, dependencies, scheduling, cron and event triggers, and stores the history of every task for debugging and replay. Workers run on your infrastructure with the engine managed or self-hosted; the project is MIT-licensed open source.
Startupkeyword match
The Enterprise AI Agent Company · United States
Sema4.ai builds an enterprise AI agent platform. Its GitHub organization ships a VS Code extension for developing AI Actions and automations in Python and infrastructure assets for self-hosted Sema4.ai deployments.
Startupkeyword match
Reach the frontier of enterprise growth
WRITER is an enterprise AI platform whose core product, WRITER Agent, executes multi-step work across business systems. It provides Playbooks (natural-language multi-step workflows), reusable Skills, Connectors for acting in business apps, AI Studio for monitoring and governing agents, and an Enterprise Brain context layer, running on its own Palmyra models.
Startupkeyword match
Run coding agents without babysitting them
MartinLoop is a runtime layer that wraps AI coding agents such as Claude Code, Codex and Gemini CLI. It lets teams hand agents larger software jobs with spend limits, required checks before results are accepted, and a record of each run.
Startupkeyword match
Run many AI coding agents in parallel from one desktop app
Baton is a desktop app for running multiple AI coding agents at once, each in its own isolated git worktree, with progress monitoring, diff review and notifications. It supports Claude Code, Codex CLI, OpenCode, Antigravity CLI, Gemini CLI, Amp, Goose and other presets, plus custom terminal agents. It is free for up to 4 workspaces, and a paid license removes the limit.
Startupkeyword match
Kanban board that hands tasks to your agent harness with persistent context
AgentKanban is a kanban board for planning work and handing tasks off to AI agents, keeping agent chat activity attached to each task. Claude Code, Codex, Cursor or any MCP-compatible agent can load a task by ID, resume it with context and update the board. It also ships a VS Code extension.
Startupkeyword match
The AI agent harness for payment teams · San Francisco, CA
Runtime is a platform for building and running AI agents across payments, risk, compliance and support workflows. Agents can run in the customer's own cloud, and sensitive steps such as releasing funds or sending replies wait for human approval. The footer says it is built in San Francisco.
Startupkeyword match
An orchestration layer for coding agents such as Claude Code and Codex, from YC-backed 21st.
1code is an open-source orchestration layer for coding agents, built by 21st, a Y Combinator W26 company that YC lists as making "UI building blocks for the agentic internet". The 1code repository describes itself as an orchestration layer for Claude Code and Codex, so several agents can be run and supervised together.
Startupkeyword match
A helpful agent and terminal multiplexer for running several coding agents side by side.
Corgi presents itself as "a helpful {agent, terminal} multiplexer" and "friendly helper for your agents": a workspace for running and switching between several terminal coding agents at once. It launched with a Show HN on 23 October 2026 that drew 141 points and 47 comments.
Startupkeyword match
A control plane for an AI software factory: governed, stateful agent workflows across the software lifecycle.
Overcut runs AI coding agents as governed, stateful workflows for code review, CI fixes, ticket triage and design-to-PR work. Teams build agents on any model (Claude, Cursor, GitHub Copilot or OpenAI agents), connect GitHub, GitLab, Bitbucket, Azure DevOps, Jira, Linear and Slack, and add approval gates, RBAC, audit trails and per-run sandboxes. It runs as SaaS, in a customer VPC or on-prem, and Gartner's Innovation Insight: AI Software Factories report (September 2026) names it as an example of a horizontal agentic software engineering platform.
Startupkeyword match
AI coding agents that let non-engineering teams ship work, and an orchestration platform for them.
Codegen built AI coding agents that let non-technical teams carry out work that used to need engineering time, and an orchestration platform for running coding agents with task context. ClickUp acquired it in December 2025 to bring coding agents into ClickUp.
Startupkeyword match
Messaging for the AI era, hosting provider for fmsg protocol
Gives people and programs an address like @user@domain.com which can then message any other fmsg address, much like email. However the fmsg protocols is more efficent, takes care of sender verification and message integrity out-of-the-box. Aiming to be the messaging layer AI agents and people use to collaborate intra and inter org.
Startupkeyword match
Background agents for engineering teams · Sofia, Bulgaria
Felan runs background AI agents for software engineering teams. They pick up the recurring work nobody should have to watch, like dependency updates, test fixes, flaky CI and small refactors. Agents start from events, schedules or a person, run on their own, and hand back a result your team reviews. They run in Felan's managed cloud or on your own infrastructure. What makes it different: it's built to keep agent costs low and visible. - No markup on tokens. Use your own API keys or subscriptions, or buy credits at provider rates. - No lock-in. Works with any popular model or provider, including local and private models, so you can use cheap models for simple tasks and top models for hard ones. - Felan Code, our open-source coding agent, is tuned to use fewer tokens without lowering quality. It's the same agent on your laptop and on the platform. - Per-task cost visibility. You see what every agent and task costs. Felan sits next to the local agents your developers already use. It doesn't replace them.
Open sourcekeyword match
LangChain/langchain · Python
LangChain provides chains, tool abstractions, integrations and runnables for composing LLM applications. It became the default on-ramp for agent development and remains among the most-starred projects in the space.
Open sourcekeyword match
OpenAI/openai-agents-python · Python
The successor to the Swarm experiments: a primitives-based SDK (agents, handoffs, guardrails, sessions) with built-in tracing and MCP support. OpenAI's main contribution to the orchestration layer.
Open sourcekeyword match
OpenAI/openai-agents-js · TypeScript
OpenAI's Agents SDK for TypeScript/JavaScript — the same primitives (agents, handoffs, guardrails) targeting voice and web runtimes, extending the SDK beyond Python.
Open sourcekeyword match
Google/adk-python · Python
ADK is Google's open-source framework for agent development: modular components, multi-agent hierarchies, tight Vertex AI integration and first-class A2A support — Google's major entry into the framework wars.
Open sourcekeyword match
AWS/harness-sdk · Python
Strands is a model-driven agent SDK where the loop is directed by the model with typed tools, and batteries included for AWS. Its Python repo was renamed to harness-sdk in 2026 — AWS now talks the harness language natively, pitching it as a way to build an agent harness and control it end-to-end, with native MCP and multi-agent composition.
Open sourcekeyword match
Microsoft/agent-framework · Python
In 2025 Microsoft began converging AutoGen and Semantic Kernel into a single Agent Framework — multi-agent workflows plus enterprise-grade process orchestration in Python and .NET. The successor to two of the earliest agent frameworks.
Open sourcekeyword match
Microsoft/semantic-kernel · C#
Semantic Kernel brought agent concepts to the enterprise .NET world — planners, plugins, memory and process frameworks — and remains widely deployed in large organizations. Now feeding into the Microsoft Agent Framework.
Open sourcekeyword match
Microsoft/autogen · Python
AutoGen pioneered agents that talk to each other (and to tools) to solve tasks. Microsoft is converging it with Semantic Kernel into the new Microsoft Agent Framework, which succeeds it.
Open sourcekeyword match
CrewAI/crewAI · Python
CrewAI popularized 'crews' — role-based multi-agent teams where each agent has a role, goal and backstory. Ships with an enterprise control plane for deploying, monitoring and governing agent crews.
Open sourcekeyword match
Agno/agno · Python
Agno (formerly Phidata) markets itself as 'the programming language for agentic software' — a fast agent runtime with built-in memory, knowledge, storage and a monitoring UI, aiming to be the production substrate beneath frameworks.
Open sourcekeyword match
LlamaIndex/llama_index · Python
LlamaIndex provides ingestion, indexing and retrieval primitives that give agents grounded context over enterprise data — plus agent and workflow abstractions. The foundational piece of the retrieval side of the harness stack.
Open sourcekeyword match
Letta/letta · Python
Letta (the company behind MemGPT) builds an agent server where agents have persistent memory, self-editing state and tool use — the reference open-source take on memory as a first-class part of the harness.
Open sourcekeyword match
Stanford NLP/dspy · Python
DSPy introduced programming models for LLM pipelines — declarative modules whose prompts and weights are optimized against metrics. Its optimizer-first philosophy reshaped how production harnesses treat prompts.
Open sourcekeyword match
Pydantic/pydantic-ai · Python
Pydantic AI brings the type-safety and validation of Pydantic to agents — typed tools, structured outputs, dependency injection and graph-based control flow, from the team behind the most-used data validation library in Python.
Open sourcekeyword match
Hugging Face/smolagents · Python
Hugging Face's smolagents keeps the agent loop tiny and lets agents write Python code as actions (CodeAct) instead of JSON tool calls. A refreshing counterpoint to heavyweight frameworks.
Open sourcekeyword match
VoltAgent/voltagent · TypeScript
VoltAgent ships a TypeScript agent framework with observability, evals and a UI out of the box — aiming to be the batteries-included option for JS/TS teams.
Open sourcekeyword match
deepset/haystack · Python
Haystack (by deepset, Berlin) is a mature framework for NLP pipelines, RAG and agents with a strong enterprise footprint in Europe — pipelines-as-graphs long before it was fashionable.
Open sourcekeyword match
The Pocket/PocketFlow · Python
PocketFlow went viral by distilling the agent loop into a hundred lines of Python. Now the scaffold beneath thousands of AI-generated agent projects — minimalism as a feature.
Open sourcekeyword match
Cloudflare/agents · TypeScript
Cloudflare's Agents SDK builds agents on Durable Objects — stateful, always-on, running at the edge with built-in scheduling, WebSockets and AI bindings. The serverless answer to agent runtimes.
Open sourcekeyword match
AgentScope (Alibaba-initiated)/agentscope · Python
AgentScope — initiated by Alibaba and now community-driven — focuses on fault-tolerant multi-agent systems with message synchronization and distributed execution, plus an increasingly 'observability-first' developer experience.
Open sourcekeyword match
Significant Gravitas/AutoGPT · Python
AutoGPT gave GPT-4 a loop, a memory and tools and let it run — kicking off the entire autonomous-agent category. Now rebuilt as a platform for building, deploying and continuously improving agents.
Open sourcekeyword match
CAMEL-AI/camel · Python
CAMEL is a research-born multi-agent framework (role-playing, tool use, data generation) that grew into one of the largest open agent-research communities — the academic counterweight to product frameworks.
Open sourcekeyword match
FoundationAgents/MetaGPT · Python
MetaGPT encodes Standard Operating Procedures into multi-agent roles (PM, architect, engineer) — the influential 'software company as a multi-agent system' experiment from 2023.
Open sourcekeyword match
VMware/Broadcom/spring-ai · Java
Spring AI brings agents, tool calling, chat memory and RAG to Spring Boot — the port of the harness layer into the world's largest enterprise Java estate. Increasingly the default for agent work in banks and insurers.
Open sourcekeyword match
LangChain4j/langchain4j · Java
LangChain4j ports the LangChain vocabulary — models, tools, memory, RAG — to idiomatic Java, and has become the main agent on-ramp for JVM shops.
Open sourcekeyword match
OpenAI/swarm · Python
Swarm was OpenAI's lightweight, experimental framework for ergonomic multi-agent orchestration. Officially superseded by the OpenAI Agents SDK, which turned its ideas into a production SDK — historically important, no longer maintained.
Open sourcekeyword match
Prefect (archived)/ControlFlow · Python
ControlFlow gave agents typed task objects and structured control flow with Prefect's orchestration underneath. Archived in 2025 as Prefect focused on its data platform and, later, FastMCP.
Open sourcekeyword match
HumanLayer/12-factor-agents · TypeScript
HumanLayer's '12-Factor Agents' distilled harness engineering into twelve principles — own your control flow, own your context window, compact state, tools as structured outputs. Required reading that shaped how the industry talks about harnesses.
Open sourcekeyword match
OpenBMB/ChatDev · Python
ChatDev simulates a software company — CEO, CTO, engineers chatting in phases — to produce whole programs. An early, much-cited multi-agent software experiment.
Open sourcekeyword match
LangChain/deepagents · Python
DeepAgents packages the patterns that make research and long-horizon agents work — planning, sub-agents, file-system memory and a tool-calling loop — into a small library that runs on any LangGraph backend. The 'just give me the harness' option from the LangChain team.
Open sourcekeyword match
ByteDance/deer-flow · Python
DeerFlow is ByteDance's open-source 'SuperAgent' harness for long-horizon work — research, coding and content creation coordinated with sandboxes, memory and report generation. One of the highest-starred agent frameworks out of China.
Open sourcekeyword match
Earendil Works/pi · TypeScript
Pi is a small, extensible agent harness: a unified multi-provider LLM API, agent runtime, TUI and coding-agent CLI, usable interactively, in print/JSON mode, over RPC or through a TypeScript SDK. It reached 1.0 in October 2026; MCP, tool search and a sandboxed JavaScript 'codemode' ship as built-in extensions, while sub-agents and plan mode are left to packages.
Open sourcekeyword match
HKUDS (HKU Data Science)/OpenHarness · Python
OpenHarness is the HKU Data Science lab's open agent harness, shipping with Ohmo — a personal agent that plans, browses and automates out of the box. The academic take on the production harness.
Open sourcekeyword match
Stacklok/mecatl · Go
Mecatl is Stacklok's open-source agent harness, architecturally separating the agent loop that reasons from the environment that executes code. The same backend runs on a laptop or a Kubernetes cluster — K8s-native from day one, not retrofitted — with TUI and web clients connecting to one engine. Early-stage but one of the few harnesses designed for platform teams.
Open sourcekeyword match
Walking Labs/learn-harness-engineering · TypeScript
A hands-on tutorial repository that teaches harness engineering from first principles — building an agent loop, tools, memory and evals step by step. The de-facto onboarding material for engineers entering the field.
Open sourcekeyword match
OpenAI/symphony · TypeScript
Symphony turns project work into isolated, autonomous coding-agent implementation runs and surfaces the ones that pass — so a team manages outcomes instead of supervising each session. OpenAI's open-source take on orchestration above the single-agent harness.
Open sourcekeyword match
ruvnet/ruflo · TypeScript
Ruflo, originally Claude Flow, is Reuven Cohen's (rUv) agent meta-harness for Claude Code and Codex. Running npx ruflo init adds about 98 agents, 60+ commands, 30 skills, an MCP server, hooks and persistent memory with swarm coordination to a project. It also installs as Claude Code plugins with a console, and supports optional federation between machines.
Open sourcekeyword match
Vercel/ai · TypeScript
The Vercel AI SDK is the React/Node toolkit for streaming, tool calls and agent-style UIs, provider-agnostic across every major model API. For TypeScript harness builders it is often the layer between the loop and the model.
Open sourcekeyword match
EverMind AI/Raven · Python
Raven is EverMind AI's 'harness of harnesses': a trusted, persistent, self-evolving multi-agent ecosystem designed for recursive self-improvement — agents that modify their own harness under governance rather than by fiat.
Open sourcekeyword match
Google/antigravity-cli · TypeScript
The Antigravity CLI brings the reasoning, execution and orchestration capabilities of Google's Antigravity platform to the command line — headless access to the same harness the IDE exposes, with a companion Python SDK (antigravity-sdk-python).
Open sourcekeyword match
AIMing Lab/AutoHarness · Python
AutoHarness (AIMing Lab) wraps any LLM client in ~2 lines for automated harness engineering: a 6–14 step pipeline, YAML constitution, risk-pattern matching, session persistence with cost tracking and multi-agent profiles. Governance-first, from academia.
Open sourcekeyword match
MaxGfeller/open-harness · TypeScript
open-harness is a code-first, composable TypeScript Agent class on the Vercel AI SDK: streaming events, filesystem/bash tools, MCP and subagent delegation — a small, readable base for rolling your own harness.
Open sourcekeyword match
Ryan Alberts/best-of-Agent-Harnesses · Markdown
Best of Agent Harnesses curates and ranks 167 harnesses by capability, autonomy and recovery — with head-to-head comparisons, templates, playbooks, an MCP server and a graveyard of dead or star-manipulated projects. The field's de-facto index.
Open sourcekeyword match
Fission AI/OpenSpec · TypeScript
OpenSpec is an MIT-licensed spec-driven development framework, built for existing codebases and not only new ones. Skills such as /openspec-propose guide a coding assistant through each change; since 2.0 they are installed once in .agents/skills/ and shared by every tool, replacing the old /opsx:* commands. A proposal creates a change folder with proposal.md, requirement and scenario specs, design.md and a tasks.md checklist, which the agent then implements. It is distributed on npm as @fission-ai/openspec and on Homebrew. It supports 30+ tools, including Claude Code, Cursor, GitHub Copilot, Codex and Kiro, and allows third-party schema bundles.
Open sourcekeyword match
code-yeongyu/oh-my-openagent · TypeScript
OmO, formerly the OpenCode edition oh-my-opencode, is a coding agent harness invoked with one omo command. It runs on senpi, its fork of the pi coding agent, and adds multi-model 'ultracode' workflow graphs triggered by the 'ulw' keyword, a memory system, CodeMode and Anthropic subscription support. Its CLI handles session resume, reasoning-effort control and an omo doctor that maps which task categories the configured providers can run. omo setup migrates provider keys, MCP servers and skills from older installs. It is licensed under the Sustainable Use License (SUL-1.0), not an OSI licence.
Open sourcekeyword match
HKUDS (Data Intelligence Lab @ HKU)/nanobot · Python
nanobot, from the HKU Data Intelligence Lab in Hong Kong, is an MIT-licensed, ultra-lightweight, self-hosted personal agent framework in Python. It runs in a browser WebUI, the terminal or chat apps: Telegram, Discord, Slack, WeChat, Feishu, Teams, email, Mattermost and Linear. Built-in tools cover shell, web search and fetch, Office files, MCP, cron and multi-agent delegation. Session history and long-term memory ('Dream') persist across sessions, and it supports long-horizon goals and scheduled automations. A Python SDK and an OpenAI-compatible API are included, and it works with hosted or local models.
Open sourcekeyword match
zhayujie/CowAgent · Python
CowAgent, previously named chatgpt-on-wechat, is an MIT-licensed personal AI assistant that calls itself a complete agent harness. Messages arrive through channels such as the web and major IM platforms. An agent core then plans and reasons over memory, a knowledge base, tools and skills, and replies through the same channel. It can control the computer and external services, create and run skills, form multi-agent teams and evolve its long-term memory. It runs 24/7 on a PC or server, deploys via a one-line installer or Docker, has a web console, and is managed with the cow CLI.
Open sourcekeyword match
AstrBotDevs/AstrBot · Python
AstrBot is an open-source, all-in-one agent chatbot platform that plugs LLM agents into mainstream instant-messaging apps, including QQ, WeChat Work, Feishu, DingTalk, Telegram and Slack. It supports agents, MCP, skills, knowledge bases, personas and automatic context compression. It can also integrate agent platforms such as Dify, Alibaba Cloud Bailian and Coze. It ships an agent sandbox for isolated code and shell execution, a WebUI and web ChatUI, and a plugin system with more than 1,000 one-click plugins. Its GitHub description calls it an OpenClaw alternative.
Open sourcekeyword match
Yeachan Heo/oh-my-claudecode · TypeScript
oh-my-claudecode (OMC) adds multi-agent orchestration to Claude Code and installs as a Claude Code plugin or an npm CLI. Since v4.1.7, Team mode is its main orchestration surface. It runs a staged pipeline (team-plan, team-prd, team-exec, team-verify, then a team-fix loop) either as an in-session native team (/team) or as tmux-launched Claude, Codex and Gemini CLI workers (omc team). OMC also ships in-session workflows such as /autopilot, /ralph and /deep-interview, plus named autopilot stage profiles configured in .claude/omc.jsonc. A sibling project, oh-my-codex, does the same for OpenAI Codex CLI.
Open sourcekeyword match
ByteDance/UI-TARS-desktop · TypeScript
TARS is ByteDance's open-source multimodal AI agent stack, and the repository ships two projects. Agent TARS is a general multimodal agent that brings GUI-agent and vision abilities to the terminal, computer, browser and product. It runs through a CLI and a Web UI and connects to real-world MCP tools. UI-TARS Desktop is a desktop app that provides a native GUI agent based on the UI-TARS model. It ships local and remote computer operators as well as browser operators.
Open sourcekeyword match
AgentScope/QwenPaw · TypeScript
QwenPaw is a personal AI assistant from the AgentScope organisation that runs locally or in the cloud and is reachable from DingTalk, Lark, WeChat, Discord, Telegram, iMessage and QQ. It has three-layer memory powered by ReMe and a built-in local runtime with QwenPaw-Flash models (2B to 35B-A3B), plus Ollama, LM Studio and 14+ cloud providers. Security features include a kernel-level sandbox, Tool Guard, File Guard, a skill scanner and access policies. It also supports sub-agents, Agent Communication Protocol (ACP) orchestration, skills, plugins and MCP. Recent releases added browser-use, computer-use and Codex/Qoder agent integration.
Open sourcekeyword match
iOfficeAI/AionUi · TypeScript
AionUi is a free, open-source 'Cowork' desktop app where AI agents work on the user's computer, reading files, writing code, browsing the web and automating tasks under user approval. It ships a built-in agent engine that needs no setup. It also runs dozens of external CLI agents in one interface, including Claude Code, Codex, Qwen Code, Hermes Agent and Cursor Agent, and its GitHub description names OpenClaw and OpenCode too. Users can reach it remotely through a WebUI, Telegram, Lark, DingTalk or WeChat, and schedule cron jobs for 24/7 unattended automation. It works with any API key and runs cross-platform.
Open sourcekeyword match
ZeroClaw Labs/zeroclaw · Rust
ZeroClaw is a personal-assistant agent runtime delivered as a single Rust binary that users configure and run on their own hardware. It talks to LLM providers including Anthropic, OpenAI, Ollama and about 20 others. It reaches users through 30+ channels such as Discord, Telegram, Matrix, email, voice, webhooks and a CLI. It acts through tools including the shell, a browser, HTTP and MCP servers. A web dashboard handles chat, memory, config and cron. Since v0.9 it can also sign in other people through your own OpenID Connect provider, each kept to a permission profile. Installs are documented for Linux, macOS, Windows, FreeBSD, NixOS and Docker.
Open sourcekeyword match
nanocoai/nanoclaw · TypeScript
NanoClaw is a personal AI assistant that runs each agent in its own Linux container, so agents get filesystem isolation instead of relying only on permission checks. Its author built it as a small alternative to OpenClaw, which the README describes as nearly half a million lines of code. NanoClaw is one process and a handful of files. It connects to WhatsApp, Telegram, Slack, Discord, Gmail and other channels, and supports memory and scheduled jobs. It runs on Anthropic's Agents SDK. Setup installs Docker, a credential gateway and the agent container. v2 can give each agent its own Slack bot identity, container and memory.
Open sourcekeyword match
Bloop AI/vibe-kanban · Rust
Vibe Kanban, from Bloop AI Limited, pairs a kanban issue board with workspaces where coding agents execute: each workspace gives an agent its own branch, terminal and dev server. It supports switching between 10+ agents (Claude Code, Codex, Gemini CLI, GitHub Copilot, Amp, Cursor, OpenCode and others), inline diff review that sends feedback to the agent, a built-in preview browser and PR creation. It installs with `npx vibe-kanban` and can be self-hosted. The website states the commercial product is sunsetting and that the project continues as open source and community maintained.
Open sourcekeyword match
Paseo/paseo · TypeScript
Paseo is an open-source agentic development environment with desktop, mobile, web and CLI clients that connect to a local daemon managing your coding agents. It runs many agents in parallel, each in its own git worktree, and lets agents create worktrees, launch and message other agents across providers. Supported agent CLIs include Claude Code, Codex, GitHub Copilot, OpenCode, Pi, Antigravity and Muse Code. It bundles editor, terminals, diffs, PRs and a browser, supports daemon-side plugins and voice control, and states it has no telemetry or forced log-ins.
Open sourcekeyword match
Microsoft/agent-lightning · Python
Agent Lightning is Microsoft's framework for training AI agents with reinforcement learning. Version 1.0 (open-sourced August 2026) uses three components: a Trainer running verl and vLLM, an API Gateway that proxies model requests and captures training data, and a Rollout Controller that runs agents locally or as Kubernetes Jobs. Agents keep their real harnesses, tools and control flow with no code changes. The project reports a Qwen3.5-9B coding-agent example improving SWE-bench Verified from 41.8% to 56.4% with 6K training samples.
Open sourcekeyword match
RightNow AI/openfang · Rust
OpenFang is an open-source Agent Operating System written in Rust and shipped as a single binary of about 32MB with a local dashboard. Its core concept is 'Hands': pre-built autonomous capability packages (manifest, system prompt, SKILL.md and guardrails such as approval gates) that run on schedules without prompting, e.g. Researcher and Lead. The site lists a WASM tool sandbox, Ed25519 manifest signing, a Merkle audit trail, SQLite plus vector memory, 40 channel adapters, and MCP and A2A support. The README marks it pre-1.0 (v0.5.10, April 2026) and warns of breaking changes.
Open sourcekeyword match
HKU Data Intelligence Lab (HKUDS)/DeepCode · Python
DeepCode, from the HKUDS lab at the University of Hong Kong, is an open-source coding agent described as combining an agent harness, loop engineering and multi-agent orchestration. It offers TUI, desktop and web clients over one shared local service that share projects, session history, models, skills, permissions, goals and automations. It emphasizes goal-driven loops, verification via tests, builds and diffs, and durable sessions that survive restarts or model switches. Its original research direction, Paper2Code, turns papers and reference repositories into working implementations; an accompanying paper is arXiv:2512.07921.
Open sourcekeyword match
lsdefine/GenericAgent · Python
GenericAgent is a minimal autonomous agent framework: about 3K lines of seed code, 9 atomic tools and a ~100-line agent loop that give an LLM system-level control of a local computer, covering browser, terminal, filesystem and keyboard/mouse. Its design principle is to evolve rather than preload skills: each solved task is crystallized into a reusable skill, building a personal skill tree over time, and the project claims about 6x lower token consumption. The README states official channels are GitHub and gaagent.ai, with DintalClaw as the sole authorized commercial partner. MIT licensed.
Open sourcekeyword match
Google/ax · Go
Agent Executor (AX) is Google's open, declarative orchestrator for running autonomous agent workloads at scale. Users declare Tasks and Workspaces (git repos, files such as AGENTS.md, MCP servers, skills) plus Model and Gateway specs in YAML; AX sandboxes each task with CPU and memory limits, wires up its workspace, enforces network policies with credential injection, and checkpoints idle agents for sub-second resume. It runs on top of the Agent Substrate runtime and targets millions of tasks per cluster. The README warns it is in heavy development with major breaking changes likely before a stable release.
Open sourcekeyword match
Orchestrator.inc/agent-orchestrator · Go
Agent Orchestrator, now branded Orchestrator.inc, is an open-source desktop workspace and CLI (`ao`) for running many coding agents at once. A project's main agent plans work and spawns workers, each with its own agent, branch and worktree; sessions sort into lanes by PR, CI and review status, and failed checks, review comments and merge conflicts are routed back to the owning worker. The site says it supports 38 coding agents, including Claude Code, Codex, Cursor, Aider, Goose and Gemini CLI. iPhone and Android apps control desktop agents; local use is free, with paid cloud workspace plans.
Open sourcekeyword match
Aden/hive · Python
Hive, by Aden, describes itself as a multi-agent harness for production AI and an open-source runtime for 'colonies' of AI agents. A long-lived 'queen' agent with a role, memory and tools does the first piece of a task itself, saves the method as a reusable skill, then fans the rest out to parallel worker agents. Results land in a shared tracker (a SQLite table) that can be queried, audited and resumed, so progress is kept on disk rather than in a context window. No workflow graph is drawn up front. Requires Python 3.11+ and Node.js 20+.
Open sourcekeyword match
Omnigent/omnigent · Python
Omnigent is an open-source meta-harness: a common orchestration layer over Claude Code, Codex, Cursor, OpenCode, Goose, Antigravity, GitHub Copilot, Devin, Kiro, Qwen Code, Kimi Code, Grok Build, Hermes, Pi and custom YAML-defined agents. Users can swap or combine harnesses without rewriting, have one agent review another's work, and continue sessions across terminal, browser, phone or a macOS desktop app. Sessions can run in disposable cloud sandboxes (Modal, Daytona, E2B, Kubernetes, Databricks and others) or on any machine over SSH, automations can start on a schedule or from Jira, Slack, GitHub, Linear or PagerDuty events, and policies govern agent behaviour. It works with API keys, Claude/ChatGPT subscriptions or compatible gateways.
Open sourcekeyword match
EvoMap/evolver · JavaScript
Evolver is EvoMap's self-evolution engine for AI agents, based on its GEP (Genome Evolution Protocol), which records evolution as auditable Genes, Capsules and Events. EvoMap's site describes GEP as letting agents share, validate and inherit capabilities across models. A linked arXiv paper (2604.15097) argues compact 'Gene' representations beat documentation-style skill packages for accumulating experience. Released open source in Feb 2026 (MIT, then GPL-3.0); the README says future releases will move to source-available, while published versions keep their original licenses.
Open sourcekeyword match
SuperPlane/superplane · Go
SuperPlane is an 'open source factory for one-shot engineering' that turns high-confidence backlog issues into verified, review-ready pull requests. It coordinates coding agents, source control, CI, review, approvals and feedback in one system, assesses which issues agents can handle, applies workflow-level guardrails to every run and sends failures back to the agent with context. An agent review scores merge confidence before human review. It runs as SuperPlane Cloud or self-hosted, with the engine under Apache-2.0 and an /ee directory under an enterprise license.
Open sourcekeyword match
mvschwarz/openrig · TypeScript
OpenRig is open-source software for running a network of coding agents as a persistent, organized team. Its pitch: 'A harness wraps a model. A rig wraps your harnesses.' Teams are defined in YAML and booted with one command, mixing Claude Code and Codex in the same rig with roles, shared context and owned work; a lead agent coordinates specialists and surfaces decisions to the user. It installs via a 'rigs' skill invoked from Claude Code or Codex, and needs Node.js 22/24, tmux and macOS or Linux (WSL2 on Windows).
Open sourcekeyword match
Osaurus/osaurus · Swift
Osaurus is an open-source AI harness for macOS built in native Swift for Apple Silicon. It sits between the user and any local or cloud model and keeps agents, memory, tools and identity on the user's machine. Each agent has its own prompts, memory and theme; tools and skills are selected via RAG, and the project lists persistent memory, autonomous execution and cryptographic identity among its features. It works fully offline with local models (an optional build bundles a ~4 GB model) and requires macOS 15.5+. Its site notes a Windows public beta.
Open sourcekeyword match
Swarms (kyegomez)/swarms · Python
Swarms is a Python framework for building and deploying multi-agent systems, with ready-made harnesses for sequential, concurrent and hierarchical architectures and graph workflows. Agents combine an LLM, tools and memory, and can run with max_loops='auto' to decide when a task is complete. It integrates with MCP, x402 and skills, has optional speedups via a drop-in LLM gateway and rustworkx graphs, and ships a Docker image. The swarms.ai site adds a hosted agent API and a marketplace for agents, prompts, tools and skills.
Open sourcekeyword match
Builderz Labs/mission-control · TypeScript
Mission Control, from Builderz Labs, is a self-hosted control plane for operating AI agents: dispatch tasks, inspect runs, review failures, track spend and coordinate agent runtimes from one local dashboard backed by SQLite. It sits above runtimes such as OpenClaw, Claude Code and Codex without replacing their reasoning or tool loops, and supports workspaces, schedulers, webhooks and alerts; live session messaging needs a connected runtime gateway. It is alpha software, runs from source (Node.js 22+, pnpm) or Docker, and includes a hardened Compose overlay.
Open sourcekeyword match
AG2/ag2 · Python
AG2 (formerly AutoGen) is an open-source programming framework for building AI agents and multi-agent cooperation, maintained by a group of volunteers from several organizations, with Chi Wang and Qingyun Wu as project administrators. As of v1.0 the protocol-driven framework is imported as `ag2`, built around an Agent class and a hub-and-channels Network for multi-agent work. The classic AutoGen-derived API (ConversableAgent, GroupChat, `import autogen`) moved to the separate ag2ai/ag2-classic repository.
Open sourcekeyword match
VRSEN/agency-swarm · Python
Agency Swarm is a Python framework for building multi-agent applications modeled on real organizational structures, started by Arsenii Shatokhin (VRSEN). Version 1.x extends the OpenAI Agents SDK and Responses API: agents get distinct roles, instructions and Pydantic-typed tools, communicate through a dedicated send_message tool along explicit, directional communication flows defined on the Agency, and can persist threads via callbacks. Non-OpenAI models are supported through LiteLLM.
Open sourcekeyword match
gptme/gptme · Python
gptme is a personal AI agent that runs anywhere a terminal does: laptops, SSH sessions, tmux, headless servers and CI. Launched in spring 2023, it was one of the first agent CLIs. It ships shell, Python, web and vision tools, supports plugins, skills and lessons, integrates with MCP and ACP, and includes guardrails. It works with Anthropic, OpenAI, Google, DeepSeek, OpenRouter and local models via Ollama, and can serve as the foundation for building your own persistent autonomous agent.
Open sourcekeyword match
evalstate/fast-agent · Python
fast-agent is a Python toolkit for interacting with LLMs that can be used as a coding agent, development toolkit, evaluation harness or workflow platform. An interactive session with shell support starts via uvx. It manages skills from a registry (LSP, hooks, compaction strategies, automation), connects to stdio or streamable-HTTP MCP servers with OAuth, and can run inside any ACP client through fast-agent-acp. It supports many providers, including Hugging Face inference providers, OpenAI Codex plans and local llama.cpp models.
Open sourcekeyword match
OneCLI/onecli · TypeScript
OneCLI is an open-source platform for running AI agents as a team. It started as a Rust credential vault for agents such as Hermes, OpenClaw and NanoClaw. Version 2 provisions one agent per employee from the company identity provider. Each agent runs in an isolated sandbox whose only exit is a gateway that injects credentials and enforces a single team policy. It supports deterministic human-in-the-loop approvals in chat, shared team-level connections and platform-kept memory. It can be self-hosted, including via Helm, or used through the hosted onecli.sh.
Open sourcekeyword match
RunMaestro/Maestro · TypeScript
Maestro is a cross-platform desktop app for running many AI coding agents and projects in parallel. It supports Claude Code, OpenAI Codex, OpenCode and Factory Droid, with Copilot CLI and others in beta, and passes each provider's existing MCP tools, skills and permissions through unchanged. Features include git-worktree sub-agents, Auto Run playbooks that execute markdown task lists in fresh sessions, moderated group chat between agents, retries on provider errors, mobile remote control, a headless CLI, and cost tracking.
Open sourcekeyword match
Mikey O'Brien/ralph-orchestrator · Rust
Ralph Orchestrator is an implementation of the 'Ralph Wiggum' technique: a hat-based orchestration framework that keeps an AI coding agent in a loop until the task is complete. After `ralph init` with a backend such as Claude, `ralph plan` runs an interactive session that writes requirements, design and implementation-plan specs under `.ralph/specs/`. `ralph run` then iterates until the agent emits LOOP_COMPLETE or hits the iteration limit. It is written in Rust and distributed via npm, Cargo and GitHub Releases.
Open sourcekeyword match
Kocoro Lab/Shannon · Go
Shannon is a multi-agent orchestration framework aimed at production use. It runs agents on Temporal workflows with time-travel replay debugging, enforces hard per-task and per-agent token budgets with model fallback, and supports human approval workflows and swarm collaboration. Code execution is sandboxed with WASI, policies use OPA, and observability comes via event streaming, Prometheus metrics and OpenTelemetry. It works with OpenAI, Anthropic, Google, DeepSeek, xAI and local models via Ollama, and exposes a REST API, Python SDK and desktop app.
Open sourcekeyword match
The Open Engine/zeroshot · Rust
zeroshot is a meta-harness for coding agents: users define a workflow graph of coding, independent review and repair steps, and the engine enforces its review gates and retry limits in code rather than relying on prompts. Each agent in the graph can use its own model and harness, including Pi, Claude Code, Codex or Copilot. The MIT-licensed CLI runs locally or self-hosted via Docker, with a hosted Zeroshot Platform also offered. The maintainers report a small model in a zeroshot loop matching a larger model on 20 ProgramBench tasks at a median 58% of its cost.
Open sourcekeyword match
Antigma Labs/ante · Rust
Ante is a terminal coding agent from Antigma Labs shipped as a single Rust executable with no runtime dependencies. It works like Claude Code or Codex without their model constraints, and can serve as the core for building custom harnesses; one settings profile can define a whole agent including a replacement system prompt. It runs offline through a managed llama.cpp engine with GGUF models, can drive other agent CLIs (Claude, Codex) in shared tmux terminals via /term, and is evaluated continuously on Terminal-Bench 2.1 across model families. It is in beta for macOS and Linux.
Open sourcekeyword match
AWS Labs/cli-agent-orchestrator · Python
CLI Agent Orchestrator (CAO), from AWS Labs, coordinates multiple AI coding CLIs so a supervisor agent can delegate work to specialist agents in parallel or in sequence. It runs a local cao-server, launches provider CLIs in isolated tmux terminal sessions, and gives the supervisor tools to coordinate workers while each agent keeps its native authentication and capabilities. Supported providers include Kiro CLI, Claude Code, Codex CLI, Antigravity CLI, Hermes, Kimi CLI, GitHub Copilot CLI, OpenCode, Oh My Pi, Cursor CLI and Grok Build CLI.
Open sourcekeyword match
Mozilla.ai/any-agent · Python
any-agent, from Mozilla.ai, provides a single Python interface for defining, running, tracing, serving and evaluating agents across frameworks such as Agno, Google ADK, LangChain, LlamaIndex, OpenAI Agents SDK and smolagents. It began as a research project comparing agent frameworks. The README states it is now in soft deprecation: its minimal common surface has graduated to the mozilla-ai-tinyagent package, and any-agent still accepts security and bug-fix PRs but has no new features planned.
Open sourcekeyword match
Mervin Praison/PraisonAI · Python
PraisonAI is a multi-agent framework for building autonomous agents that research, plan, code and execute tasks, from a single agent up to multi-agent teams that hand off work and verify results. The lightweight praisonaiagents core SDK creates an agent in a few lines of Python, and the project advertises built-in memory, RAG and support for 100+ LLMs. Documented use cases include research, code generation, data pipelines, workflow automation and support bots on Telegram, Discord and Slack.
Open sourcekeyword match
Happier/happier · TypeScript
Happier is an open-source, self-hostable web, desktop and mobile client and orchestrator for coding agents including Claude Code, Codex, OpenCode, Pi, Cursor and others, with any ACP-speaking agent addable from settings. Agents run on the user's own machines through a CLI and daemon, and sessions can spawn subagents across agents, for example Claude asking Codex for a review, or switch agents mid-session. Users can approve, steer, review and commit from phone, desktop or browser, with traffic end-to-end encrypted. It uses users' existing subscriptions.
Open sourcekeyword match
zeronsh/zeron · Rust
Zeron is a native control plane for coding agents, including Claude Code, Codex, Cursor, Devin, Grok, Hermes, Pi and Antigravity. It ships as a desktop app for macOS, Windows and Linux that needs no account or network connection, keeping sessions on the device, plus a headless Linux engine that runs as a background service so agents keep running on a server. Optional multi-device sync lets users start an agent on one device and follow or drive it from another. It is written in Rust and MIT licensed.
Open sourcekeyword match
Automaze/ccpm · Shell
CCPM is a project-management system for AI coding agents, now packaged as an Agent Skill usable in skills-compatible harnesses such as Claude Code, Codex, OpenCode, Factory, Amp and Cursor. It turns ideas into PRDs, PRDs into epics, epics into GitHub issues and issues into code, keeping traceability throughout. GitHub Issues serve as the shared source of truth, and git worktrees let multiple agents work on tasks in parallel without conflicting. It is built by Automaze and MIT licensed.
Open sourcekeyword match
Leon AI/leon · TypeScript
Leon is an open-source personal AI assistant first released in 2019 as an intent-classification assistant and now being rebuilt around tools, context, layered memory and agentic execution in a 2.0 Developer Preview. It offers smart, controlled (deterministic native skills) and agent (step-by-step planning) modes, supports SKILL.md-backed agent skills, and includes progressively loaded computer use for desktop and browser interfaces. It can use local or remote AI providers to balance privacy and capability.
Open sourcekeyword match
NextLevelBuilder/goclaw · Go
GoClaw is a multi-agent AI gateway shipped as a single static Go binary, originally inspired by the OpenClaw architecture. It runs an eight-stage agent pipeline (context, history, prompt, think, act, observe, memory, summarize), three-tier memory with a knowledge graph, agent teams with shared task boards and delegation, 20+ LLM providers and seven messaging channels. A multi-tenant PostgreSQL mode adds per-user workspaces, RBAC and encrypted keys; a desktop 'Lite' edition runs on SQLite. Note: licensed CC BY-NC 4.0 (non-commercial).
Open sourcekeyword match
ValueCell/ClawX · TypeScript
ClawX is an open-source desktop app that wraps the official OpenClaw core in a graphical interface, so users can run OpenClaw agents without the terminal. It provides a guided setup wizard, visual configuration with validation, automatic gateway lifecycle management, a unified provider panel and local-first skill management with an optional marketplace. It supports Windows and multiple UI languages; ValueCell also offers an enterprise edition.
Open sourcekeyword match
Langflow/langflow · Python
Langflow is an open-source platform for building and deploying AI agents and workflows. It combines a visual builder with Python-level access to customize any component, an interactive playground for step-by-step testing, multi-agent orchestration with conversation management and retrieval, and deployment as an API, exported JSON or an MCP server that exposes flows as tools. It integrates with LangSmith, Langfuse and other observability tools, and ships a desktop app for Windows and macOS.
Open sourcekeyword match
FoundationAgents/OpenManus · Python
OpenManus is an open-source general-purpose agent framework created by members of the MetaGPT team as an open alternative to Manus; its first prototype was built in three hours. It is a Python project installed via conda or uv, and starts Browser Use's CLI as a default MCP server for browser tools, with optional Browser Use Cloud remote browsers. The team also maintains OpenManus-RL, a companion project for reinforcement-learning-based tuning of LLM agents developed with UIUC researchers.
Open sourcekeyword match
Multica/multica · Go
Multica is a source-available, self-hostable workspace for assigning issues to AI coding agents as if they were teammates. Agents pick up issues, comment progress, raise blockers and hand results back for human review, while the intent, run, decisions and diff stay linked to the same issue. It works with existing agent CLIs including Claude Code, Codex, Cursor, Copilot, Kimi and OpenCode, each configured with a name, provider and runtime. It is offered self-hosted or as a cloud service, under a license based on Apache 2.0 with additional conditions.
Open sourcekeyword match
Yeachan Heo/oh-my-codex · TypeScript
oh-my-codex (OMX) is a workflow layer for OpenAI's Codex CLI that keeps Codex as the execution engine while adding stronger default sessions, a consistent clarification-to-completion workflow and invocable workflows such as $plan, $ultragoal, $team, $code-review and $ultraqa. Project guidance, plans, logs and state are kept in a .omx/ directory. It also offers an SDK-based path for integrations such as OpenClaw and Hermes. It is tuned primarily for macOS and Linux; native Windows is not the default experience.
Open sourcekeyword match
Block/buzz · Rust
Buzz is a self-hostable workspace from Block where humans and AI agents share the same rooms. It is built as a Nostr relay, so every message, workflow step, review approval and git event is a signed event in one log with a common identity model and audit trail. Agents get their own keys and channel memberships and can open repos, send patches, review code, run workflows and orchestrate other agents. Its agent surface includes buzz-cli (JSON in/out), buzz-acp (an ACP-to-MCP harness for Goose, Codex and Claude Code) and buzz-agent.
Open sourcekeyword match
Gas Town Hall/gastown · Go
Gas Town is a workspace manager for coordinating many AI coding agents (Claude Code, GitHub Copilot, Codex, Gemini and others) on different tasks. It persists work state in git-backed hooks and a Beads ledger so context survives agent restarts, and provides mailboxes, identities and handoffs between agents. Its model has a 'Mayor' AI coordinator over a town workspace containing per-project rigs with crew workspaces and 'polecat' worker agents on git worktrees; the README claims comfortable scaling to 20-30 agents.
Open sourcekeyword match
coleam00/Archon · TypeScript
Archon lets teams describe how work gets done as YAML workflows — graphs of nodes where agents supply judgment, scripts compute and humans approve — and runs them with AI agents. Its bundled SDLC pack takes an issue to a merged pull request with plan, implement-until-tests-pass loops, approval gates and PR steps. Archon is not itself a coding agent: it drives Claude Code, Codex and Pi via their CLIs and SDKs, logs events, transcripts and per-run cost, and self-hosts on SQLite or PostgreSQL.
Open sourcekeyword match
elizaOS/eliza · TypeScript
elizaOS is an open-source TypeScript framework and product stack for autonomous AI agents. Its monorepo holds the core runtime, the Eliza app, a CLI, cloud services, native bridges and first-party plugins, with Linux and Android distribution tooling for running a whole device as elizaOS. Developers build agents and plugins against the runtime, can run local inference, and use a bundled benchmark orchestrator; a hosted Eliza Cloud is also offered.
Open sourcekeyword match
Emcie/parlant · Python
Parlant is an Apache-2.0 Python platform for customer-facing chat agents in support, sales and lead capture. Since version 4 you describe how the agent should behave and your coding agent (such as Claude Code) implements it as policies, customer variables and tools. Parlant includes a chat widget, human handoff, testing with simulated customers, and a Dashboard with a matching CLI for sessions, reasoning and per-turn model costs. You bring your own model provider, and managed hosting is offered for a flat monthly fee.
Open sourcekeyword match
LangChain/open-swe · Python
Open SWE is an open-source software factory built on LangChain's Deep Agents and LangGraph. It investigates a codebase, implements changes in persistent per-thread Linux sandboxes, validates them and delivers pull requests; it also reviews PRs, learns repository review preferences, monitors CI and can be triggered from a dashboard, GitHub issues and PR conversations, Slack or Linear. The README notes it is under active development, not accepting external contributions, and that deploying as-is requires Managed Deep Agents.
Open sourcekeyword match
LastMile AI/mcp-agent · Python
mcp-agent is a Python framework built on the view that MCP is all you need to build agents and that simple patterns beat complex architectures. It fully implements MCP and manages server connection lifecycles, implements the patterns from Anthropic's Building Effective Agents as composable workflows, and can scale to durable workflows on Temporal that pause, resume and recover without API changes. Agents declare which MCP servers they use and attach an LLM provider such as OpenAI.
Open sourcekeyword match
smtg-ai/claude-squad · Go
Claude Squad is a Go terminal app that manages multiple local coding agents — Claude Code, Codex, Gemini CLI, Aider and others — in separate workspaces so several tasks can run at once. Each task gets its own isolated git workspace to avoid conflicts, tasks can run in the background including auto-accept mode, and changes can be reviewed and checked out before pushing. It installs as `cs` via Homebrew or an install script and requires tmux and gh.
Open sourcekeyword match
Astro (withastro)/flue · TypeScript
Flue is a TypeScript framework for building autonomous agents in the style of Claude Code and Codex rather than scripted LLM calls. Its built-in harness gives any model sessions, tools, skills (SKILL.md files), instructions, filesystem access and a sandbox — virtual, local or remote container — composed with hooks such as useModel, useSandbox, useSkill and useTool. Agents can run locally via CLI or be deployed. The repository lives under the withastro GitHub organization.
Open sourcekeyword match
Docker/docker-agent · Go
Docker Agent is an open-source agent builder and runtime from Docker Engineering, shipped as a docker CLI plugin (`docker agent run`). Agents are defined declaratively in YAML or HCL with a model, instructions, toolsets and sub-agents, and the runtime handles the LLM loop, tool calls and streaming. It supports built-in tools plus any MCP server, multi-agent delegation, RAG, and providers including OpenAI, Anthropic, Gemini, Bedrock and Docker Model Runner. Agents can be pushed to and pulled from OCI registries, and run as a TUI, headless CLI, HTTP/MCP server or over A2A.
Open sourcekeyword match
Sipeed/picoclaw · Go
PicoClaw is an independent open-source personal AI assistant initiated by hardware maker Sipeed and written from scratch in Go (the README states it is not a fork of OpenClaw or NanoBot, though inspired by NanoBot). It ships as a single self-hosted binary targeting under ~10-20MB of RAM, runs on Raspberry Pi, RISC-V, ARM64, x86_64, Android and Docker, connects to 16+ chat platforms such as Telegram, Discord and Slack, and supports multiple LLM providers. Recent releases added MCP server management, hooks and isolation support. The maintainers warn it is pre-1.0 and not yet production-ready.
Open sourcekeyword match
TinyAGI/tinyagi · TypeScript
TinyAGI (formerly TinyClaw) describes itself as an agent-teams orchestrator for a "one person company". It runs multiple isolated agents with specialized roles that hand off work through chain execution and fan-out, reachable over Discord, WhatsApp and Telegram. It drives the Claude Code and Codex CLIs or any OpenAI/Anthropic-compatible endpoint, persists sessions across restarts, uses a SQLite queue with retries and dead-letter handling, and includes a web portal (TinyOffice), a live TUI dashboard, team chat rooms and a plugin system for message hooks.
Open sourcekeyword match
NullClaw/nullclaw · Zig
NullClaw is an autonomous AI assistant runtime written in Zig and compiled to a tiny static binary (678 KB, ~1 MB RAM, millisecond startup per its README) aimed at cheap ARM, x86 and RISC-V boards. It claims 50+ providers, 19 channels, 35+ tools, multiple memory engines, MCP, subagents and voice, with core systems exposed as swappable vtable interfaces. Security features include pairing, sandboxing via landlock, firejail, bubblewrap or Docker, allowlists, workspace scoping and encrypted secrets. It sits within a wider Null ecosystem (NullHub UI, nullboiler orchestration, nullwatch observability).
Open sourcekeyword match
SwarmClaw/swarmclaw · TypeScript
SwarmClaw is an open-source, self-hosted AI agent runtime and multi-agent framework for running agent swarms and orchestrators. It provides heartbeats, cron schedules and webhook triggers, delegation between agents or to CLI backends (Claude Code, Codex, OpenCode, Gemini, Goose and others), agent memory, runtime skills, a Kanban task board with audit trail, and connectors for Discord, Slack, Telegram and WhatsApp. It supports 23+ LLM providers and can sit on top of OpenClaw gateways and Hermes agents. Distributed as a desktop app, Docker Compose stack or from source.
Open sourcekeyword match
Pacific Studio/openase · Go
OpenASE is an open-source, ticket-driven software-engineering platform from Pacific Studio, shipped as a single Go binary bundling an API server, workflow orchestrator and embedded web UI. Every unit of work is a ticket with a workflow; an orchestrator detects pickup statuses, a CLI agent claims the ticket, executes it on a registered machine and records every step in an activity stream. It supports Claude Code and Codex (recommended) and Gemini CLI, uses Markdown "harness" documents to constrain agent behaviour, and injects reusable skills into each CLI's skill directory. By default it launches agents with permissive flags such as --dangerously-skip-permissions.
Open sourcekeyword match
Google Cloud/scion · Go
Scion is an open-source orchestration platform from Google Cloud for teams of AI agents and the people working with them. It runs existing harnesses such as Claude Code, Antigravity, Codex and OpenCode as independent processes with per-agent configuration and credentials, in shared workspaces or separate Git worktrees, and lets agents delegate tasks, message each other and work in parallel locally or on cloud infrastructure. It offers a CLI, a Workstation web UI and a hosted Hub, a Skill Bank for versioned skills, harness configuration bundles and an A2A bridge. Agents learn the scion CLI through skills.
Open sourcekeyword match
General Action/emdash · TypeScript
Emdash, from YC W26 company General Action, is a free, open-source desktop app for macOS, Windows and Linux that runs multiple coding agents at once, each in its own Git worktree with its own branch, terminal and conversation. It is provider-agnostic (Claude Code, Codex, Cursor, Amp, Antigravity and 25+ others), pulls issues from Linear, Jira, GitHub and others, schedules recurring agent work, includes an in-app browser, diff review, PR creation and CI monitoring, connects to remote machines over SSH, and syncs prompts, skills and MCP servers across agents.
Open sourcekeyword match
T3 Tools/t3code · TypeScript
T3 Code, made by T3 Tools Inc., is an MIT-licensed GUI that drives existing coding-agent CLIs (Claude Code, Codex CLI, Antigravity, OpenCode, Cursor, Grok CLI) from a single app, using the user's own subscriptions rather than reselling tokens. Each agent thread writes to its own branch; inline diff review and a one-click commit/push/open-PR flow with generated descriptions close the loop. It runs on desktop with iOS and Android companion apps and is designed to be forked and customised.
Open sourcekeyword match
Happy Engineering/happy · Swift
Happy is a free, MIT-licensed app that orchestrates multiple coding agents (Claude Code, Codex, Grok, Gemini via Antigravity and others) from one interface, reusing subscriptions users already pay for. It ships desktop apps for macOS, Windows and Linux, iOS and Android apps with end-to-end encrypted session sync, and an npm CLI (e.g. `happy claude`, `happy codex`) whose sessions can be resumed from a phone. A team server adds multiplayer sign-in and shared projects and sessions; tasks can be delegated to subagents.
Open sourcekeyword match
coolLabs/jean · TypeScript
Jean, from coolLabs Solutions Kft., is a free Apache-2.0 workspace that combines git worktrees, chat sessions, terminals and GitHub/Linear integrations for running AI coding agents in parallel. Each task gets its own worktree; built-in flows automate review, commits, PRs and AI-assisted merge-conflict resolution, linked projects share context across repositories, and a headless jean-server lets remote Linux machines appear alongside local projects. It supports Claude CLI, Codex CLI, OpenCode, Cursor, Grok and (beta) Pi, Command Code and Kimi Code with bring-your-own subscriptions.
Open sourcekeyword match
Monorepo Labs/dray · TypeScript
Dray is an Apache-2.0 desktop app (Tauri with a React frontend and Rust backend) that wraps coding-agent CLIs such as Claude Code, Codex, fx and pi in a native chat UI. Many sessions can run at once, each with its own git worktree, diff view, PR panel, issue panel and embedded browser, and a `dray` CLI lets agents fan work out to further sessions. The project lists Vercel CEO Guillermo Rauch's OSS Grants programme and Greptile as sponsors.
Open sourcekeyword match
Nimbalyst/nimbalyst · TypeScript
Nimbalyst (the successor to the Crystal parallel-worktree app) is an MIT-licensed desktop workspace for macOS, Windows and Linux, with iOS/Android companions, for running multiple coding agents in parallel and editing their work visually across markdown, mockups, diagrams, docs and tasks. It works with existing Claude Code or Codex subscriptions or API keys. A separate Nimbalyst Wiki plugin lets MCP-compatible agents maintain a shared, human-editable wiki of decisions and open questions, with a paid Wiki Teams plan (free in beta).
Open sourcekeyword match
Imbue/sculptor · Python
Sculptor is an MIT-licensed desktop app from Imbue (an experimental research preview) for running coding agents in parallel. Users connect a repo, create workspaces (isolated worktree copies of the code), prompt agents and merge reviewed changes back. It ships with integrated support for Claude Code and the Pi agent harness, can run any terminal-based agent, and bundles workflow skills such as spec, mocks and a fix-bug flow that writes failing tests before fixing. Downloads are offered for macOS (Apple Silicon) and Linux.
Open sourcekeyword match
Omnara/omnara · Go
Omnara (YC S25) is an Apache-2.0, self-hostable API for running agents in production, positioned as an alternative to Claude Managed Agents and OpenAI's Agents API. State lives in an event log so agents are durable, recover from failures and idle without compute. Agents run on Omnara machines, the user's Blaxel, Daytona or Unikraft accounts, or an outbound-connecting daemon. It adds human approval for risky tool calls, scoped secrets, RBAC and append-only audit history, exposes REST, TypeScript SDK and CLI, and works with OpenAI, Anthropic, OpenRouter, LiteLLM or self-hosted models.
Open sourcekeyword match
Pythagora-io/gpt-pilot · Python
GPT Pilot is Pythagora's open-source AI developer that builds apps step by step with a human in the loop. The repo is maintained by Pythagora-io and released under the Functional Source License 1.1 (MIT future license).
Open sourcekeyword match
eyaltoledano/claude-task-master · JavaScript
Taskmaster is an AI-powered task-management system that plugs into coding-agent environments such as Cursor, Windsurf and Roo. The project homepage is tryhamster.com. It is licensed MIT with the Commons Clause.
Open sourcekeyword match
bmad-code-org/BMAD-METHOD · Python
BMAD Method (Breakthrough Method for Agile AI Driven Development) is an open-source framework from BMad Code, LLC. It structures AI-driven development as agile, agent-based workflows. It is MIT licensed.
Open sourcekeyword match
kbwo/ccmanager · TypeScript
CCManager is a session manager for CLI coding agents, including Claude Code, Gemini CLI, Codex CLI, Cursor Agent, Copilot CLI, Cline CLI, OpenCode and Kimi CLI. It is MIT licensed.
Open sourcekeyword match
CodebuffAI/freebuff · TypeScript
Freebuff gives you specialized coding agents and a choice of models across a CLI, a desktop app for running parallel agents, a web app builder, a cloud product that runs agents on GitHub repos, and a chat app. It needs no subscription, credits or API key, and text ads pay for the included models. The repo was formerly CodebuffAI/codebuff.
Open sourcekeyword match
just-every/code · Rust
Every Code is a fork of the OpenAI Codex CLI. It adds validation, automation, browser integration, multi-agent features and theming. It can orchestrate agents from OpenAI, Claude, Gemini or any provider.
Open sourcekeyword match
asheshgoplani/agent-deck · Go
A terminal session manager for AI coding agents: one TUI to manage Claude, Gemini, OpenCode, Codex and other agent sessions.
Open sourcekeyword match
snarktank/ralph · TypeScript
Ralph runs an AI coding tool (Amp by default, or Claude Code) in a loop until every PRD item is complete. Each iteration starts a fresh instance with clean context. It is based on Geoffrey Huntley's Ralph pattern.
Open sourcekeyword match
Dicklesworthstone/mcp_agent_mail_rust · Rust
A Rust MCP server for coordinating multiple coding agents. It has 45 tools and 25 resources, a Git-backed archive, SQLite indexing, advisory file locks, a TUI, a web UI and an agent-first CLI. It is the maintained rewrite of the original Python mcp_agent_mail, which is now deprecated.
Open sourcekeyword match
Dicklesworthstone/ntm · Go
A Go control plane that spawns, tiles and coordinates multiple AI coding agents (Claude, Codex, Antigravity, omp, Grok) across tmux panes. It adds graph-aware work triage, a safety policy with approvals, Agent Mail coordination, and checkpoints and timelines. It exposes robot JSON plus REST, SSE, WebSocket and OpenAPI interfaces.
Open sourcekeyword match
Dicklesworthstone/claude_code_agent_farm · Shell
An orchestration framework for running 20+ Claude Code agents in parallel. It is used for automated bug fixing and best-practice sweeps, with lock-based coordination and real-time tmux monitoring.
Open sourcekeyword match
moazbuilds/CodeMachine-CLI · TypeScript
CodeMachine is an open-source orchestration layer that runs AI coding CLIs (Claude Code, Codex, Cursor and others) through structured workflows, using their headless scripting modes. You define a workflow once, and it handles execution, context passing and agent coordination. Its README lists multi-agent orchestration, parallel execution and long-running workflows that last hours or days. Installed with npm i -g codemachine.
Open sourcekeyword match
AndyMik90/Aperant · TypeScript
Aperant, formerly Auto Claude, is an autonomous multi-agent coding framework that plans, builds and validates software, with a kanban-board desktop app. The 2.x desktop app is in maintenance mode while Aperant 3.0 is rebuilt from scratch with added cloud features. The README says 3.0 will stay open source under AGPL-3.0.
Open sourcekeyword match
AutoMaker-Org/automaker · TypeScript
Automaker orchestrates AI agents to build features on their own: you add features to a board, and moving one to In Progress assigns an agent to implement it. Agents read files, write code, run commands and tests, and make git commits in isolated git worktrees. It is built on the Claude Agent SDK and also supports Codex, Copilot, Cursor, Gemini and OpenCode providers. As of November 2026 its README says the project is no longer actively maintained, and the automaker.app site is gone.
Open sourcekeyword match
agent0ai/agent-zero · Python
Agent Zero is an open agent framework that gives the agent a Dockerized Linux desktop, a browser with DOM annotation, live document cowork, projects, skills and plugins. It supports per-project memory, multi-agent delegation to subagents, a plugin hub, and a CLI bridge to the host machine for working in local repositories. It runs wherever Docker runs.
Open sourcekeyword match
kortix-ai/suna · TypeScript
Suna is Kortix's open-source agent platform, described by the repo as 'The open-source AI Operating System'. It lets you self-host a general-purpose agent, and the project's homepage is kortix.com.
Open sourcekeyword match
camel-ai/owl · Python
OWL (Optimized Workforce Learning) is CAMEL-AI's multi-agent system for general assistance in real-world task automation. It is built on top of the CAMEL framework.
Open sourcekeyword match
Ido-Levi/Hephaestus · Python
Hephaestus is a semi-structured agentic framework. Its workflows build themselves as agents discover what needs to be done, instead of following a plan fixed up front.
Open sourcekeyword match
phodal/auto-dev · Kotlin
AutoDev Xiuper (AutoDev 3.0, alpha) is an AI-native, multi-agent development platform built on Kotlin Multiplatform. It covers document research, coding, code review, data query, artifact generation and web interaction. It ships as an IntelliJ IDEA plugin, VS Code extension, CLI (@xiuper/cli), web, desktop and Android apps, plus a Ktor-based remote coding agent server. A shared core provides the agent engine, tools, MCP integration and AGENTS.md loading.
Open sourcekeyword match
gotalab/cc-sdd · TypeScript
cc-sdd describes itself as a minimal, adaptable spec-driven development (SDD) harness built on Agent Skills. It turns approved specs into long-running autonomous implementation. It supports Claude Code, Codex, Cursor, Copilot, Windsurf, OpenCode, Gemini CLI and Antigravity.
Open sourcekeyword match
BeehiveInnovations/pal-mcp-server · Python
PAL MCP Server, formerly Zen MCP, is an MCP server that connects Claude Code, Gemini CLI and Codex CLI to other model providers. These include Gemini, OpenAI, OpenRouter, Azure, Grok, Ollama and custom models. It lets a coding agent consult other models within one workflow.
Open sourcekeyword match
Fosowl/agenticSeek · Python
The repo describes itself as a fully local Manus AI with no APIs or monthly bills. It is an autonomous agent that thinks, browses the web and writes code, costing only electricity to run.
Open sourcekeyword match
parcadei/Continuous-Claude-v3 · Python
Context management for Claude Code: hooks keep state across sessions using ledgers and handoffs. It also runs MCP tools without polluting the context, and orchestrates agents in isolated context windows.
Open sourcekeyword match
Th0rgal/open-ralph-wiggum · TypeScript
A CLI where typing `ralph "prompt"` starts a coding agent in a Ralph loop. It also accepts a prompt file and has a status check. It supports Open Code, Claude Code, Codex and Copilot.
Open sourcekeyword match
microsoft/amplifier · Python
Amplifier is an AI-powered modular development assistant from Microsoft, in early preview. Its README calls it a research demonstrator. The CLI is one reference interface on top of a modular, extensible platform. An earlier Claude Code-based version lives on the amplifier-claude branch.
Open sourcekeyword match
Eigenwise/atomic-agents · Python
Atomic Agents is a lightweight, modular Python framework for building AI agents out of small, composable, schema-typed components. The project is now hosted under the Eigenwise GitHub organization and is MIT licensed.
Open sourcekeyword match
julep-ai/julep · Python
Julep provides durable, composable AI agent flows that can crash and resume, retry safely, and explain every step. It is Apache-2.0 licensed and written mainly in Python.
Open sourcekeyword match
0xPlaygrounds/rig · Rust
Rig is a Rust library for building modular and scalable LLM applications and agents. It is MIT licensed and documented at rig.rs.
Open sourcekeyword match
apache/burr · Python
Apache Burr is a framework for building applications that make decisions, such as chatbots, agents and simulations. It lets you monitor, trace, persist and execute them on your own infrastructure.
Open sourcekeyword match
langroid/langroid · Python
Langroid is a Python framework for building LLM applications through multi-agent programming. It is MIT licensed and actively maintained.
Open sourcekeyword match
griptape-ai/griptape · Python
Griptape is a modular Python framework for AI agents and workflows, with chain-of-thought reasoning, tools and memory. It is Apache-2.0 licensed and backed by Griptape (griptape.ai).
Open sourcekeyword match
iii-hq/iii · Rust
iii (by Motia LLC, formerly the Motia project) turns backend concerns such as queues, cron, HTTP, state, observability, agents and sandboxes into interoperable Workers, Triggers and Functions on one live runtime. Agents can add workers, discover their functions, call them and trace results. Its SPDX file declares Apache-2.0 AND Elastic-2.0.
Open sourcekeyword match
i-am-bee/beeai-framework · Python
BeeAI Framework is for building production-ready AI agents in both Python and TypeScript. It comes from the i-am-bee organization, which also maintains the Agent Communication Protocol (ACP).
Open sourcekeyword match
genkit-ai/genkit · TypeScript
Genkit is an open-source framework for building agentic apps in JavaScript, Go, Dart and Python. It is built and used in production by Google and is Apache-2.0 licensed.
Open sourcekeyword match
cloudwego/eino · Go
Eino is an LLM and AI application development framework for Go from the CloudWeGo project. It is Apache-2.0 licensed and documented on cloudwego.io.
Open sourcekeyword match
QwenLM/Qwen-Agent · Python
Agent framework and applications built upon Qwen>=3.0. Features function calling, MCP, a code interpreter, RAG and a Chrome extension.
Open sourcekeyword match
2FastLabs/agent-squad · Python
Multi-agent orchestration framework with classifier-based routing between agents. Python and TypeScript have feature parity, and a Swift runtime runs orchestration entirely on device. Supports native and MCP tools, realtime voice and tracing. Previously hosted at awslabs/agent-squad and formerly named multi-agent-orchestrator.
Open sourcekeyword match
ogx-ai/ogx · Python
Drop-in replacement for the OpenAI API that runs on a laptop, in a datacenter or in the cloud, with a pluggable provider architecture (Ollama, vLLM, managed services). Its Responses API handles agentic orchestration on the server side, with tool calling, MCP server integration and built-in file search (RAG). Formerly Llama Stack.
Open sourcekeyword match
JetBrains/koog · Kotlin
Koog is a JVM (Java and Kotlin) framework for building predictable, fault-tolerant and enterprise-ready AI agents. It targets backend services, Android, iOS, JVM and in-browser environments. Built by JetBrains.
Open sourcekeyword match
Upsonic/Upsonic · Python
Python framework for building autonomous AI agents, with MCP tool integration for connecting agents to external data sources and services. Released under the MIT License.
Open sourcekeyword match
kaiban-ai/KaibanJS · TypeScript
JavaScript-native framework for building and managing multi-agent systems, organized around a Kanban-inspired approach to tasks.
Open sourcekeyword match
Agenta-AI/agenta · TypeScript
Open-source workspace where you build specialized agents by chatting with them. You connect apps through MCP or Composio and improve agents through feedback. Agents can be used in the web app, Slack, Telegram or WhatsApp Business, and can run as scheduled or event-triggered automations with per-run visibility. The README states it is MIT-licensed.
Open sourcekeyword match
oracle/agent-spec · Python
Open Agent Spec is a framework-agnostic declarative language for defining agentic systems. It defines building blocks for standalone agents and structured agentic workflows, plus common ways to compose them into multi-agent systems.
Open sourcekeyword match
microsoft/magentic-ui · Python
MagenticLite, formerly Magentic-UI, is an experimental agent from Microsoft that works across the browser and the local file system. The repo's homepage links to a Microsoft Research blog post on MagenticLite, MagenticBrain and Fara1.5, which describes an agentic experience optimized for small models.
Open sourcekeyword match
n8n-io/n8n · TypeScript
n8n is a fair-code workflow automation platform with native AI capabilities. It combines visual building with custom code, can be self-hosted or run in the cloud, and offers 400+ integrations.
Open sourcekeyword match
stoneforge-ai/stoneforge · TypeScript
Stoneforge is a web dashboard and runtime for orchestrating AI coding agents in parallel. Each worker runs in its own git worktree. A dispatch daemon assigns tasks once their dependencies clear, and a merge steward runs tests and squash-merges work that passes. It supports Claude Code, OpenCode and OpenAI Codex.
Open sourcekeyword match
sipyourdrink-ltd/bernstein · Python
Bernstein is an open-source governance layer for AI coding agents. It runs CLI agents in parallel git worktrees and merges only diffs that pass lint, type and test gates. A deterministic scheduler decides what runs, with no model in that loop. It enforces policy as code and keeps an HMAC-chained audit log and Ed25519-signed run receipts. The site cites 40+ agent adapters, including Claude Code, Codex and Gemini CLI, and says it is beta software built by one maintainer.
Open sourcekeyword match
compozy/compozy · Go
CompozyOS is a local runtime, a daemon with a CLI, HTTP/SSE and a web UI. It turns agent CLIs such as Claude Code, Codex, Gemini CLI and Cursor into a team that splits work, hands off tasks, runs scheduled jobs and loops, and shares one project memory. The runtime enforces task leases, permissions and approvals. It is currently in beta.
Open sourcekeyword match
microsoft/conductor · Python
Conductor is a CLI from Microsoft for defining and running multi-agent workflows with the GitHub Copilot SDK and Anthropic Agents SDK.
Open sourcekeyword match
openprose/prose · TypeScript
OpenProse is an open-source language for writing programs that AI agents run, described as 'a new kind of language for a new kind of computer'. The repository is MIT-licensed and written in TypeScript.
Open sourcekeyword match
hex/claude-council · Shell
Claude Code plugin that asks several AI coding agents the same question and shows their answers side by side. It supports Gemini, OpenAI, Grok, Perplexity, Kimi and OpenRouter models over API; codex, antigravity, grok, kimi and cursor CLIs on subscription auth; and ollama locally. Codex specialists can take coding tasks on their own branch (experimental).
Open sourcekeyword match
superhq-ai/superhq · Rust
A Rust/GPUI desktop app for macOS that runs multiple AI coding agents (Claude Code, OpenAI Codex, Pi) in isolated sandbox VMs with full terminal access. Agents never see the real API keys: an auth gateway on the host injects credentials into outgoing requests. The README describes it as very early alpha.
Open sourcekeyword match
ryanmac/code-conductor · Python
Runs multiple Claude Code sub-agents in parallel, each in an isolated git worktree, to avoid merge conflicts. Agents claim tasks, implement them and ship on their own, using GitHub Actions workflows with built-in authentication. A one-command installer detects the project stack.
Open sourcekeyword match
raine/workmux · Rust
workmux pairs a git worktree with a tmux window for each task, so several coding agents can work in parallel without touching each other's checkout. Each worktree starts its configured agent (Claude Code, OpenCode, Codex, Copilot CLI, Pi, Gemini CLI, Grok and others), and WezTerm, kitty and Zellij are supported besides tmux.
Open sourcekeyword match
revfactory/harness · Markdown
Harness is a team-architecture factory for Claude Code. Ask it to build a harness for a project, and it turns your domain description into a team of specialised agents plus the skills they use. Install it through the plugin marketplace.
Open sourcekeyword match
openai/codex-plugin-cc · JavaScript
OpenAI's official plugin for using Codex from inside Claude Code to review code or delegate tasks.
Open sourcekeyword match
gemini-cli-extensions/conductor · Python
A plugin for spec-driven development with AI coding agents: specify, plan and implement software features.
Open sourcekeyword match
rohitg00/pro-workflow · JavaScript
A Claude Code workflow layer that learns from your corrections, with self-correcting memory, context engineering, parallel worktrees, agent teams and 17 skills.
Open sourcekeyword match
agent-sh/agentsys · JavaScript
AI writes code; AgentSys automates everything else with 24 plugins, 49 agents and 44 skills for Claude Code, OpenCode, Codex, Cursor and Kiro.
Open sourcekeyword match
alvinunreal/oh-my-opencode-slim · TypeScript
Lean, fine-tuned OpenCode multi-agent suite: mix any models and auto-delegate tasks.
Open sourcekeyword match
rlaope/oh-my-hermes · Python
All-in-one operating layer for Nous Research's Hermes Agent: per-lane model chains, /dag-ulw parallel graph runs and expert skills, without patching Hermes.
Open sourcekeyword match
first-fluke/oh-my-agent · TypeScript
Load one skill or install domain presets, project hooks and parallel workflows with test gates across Claude Code, Codex, Cursor and others.
Open sourcekeyword match
nyldn/claude-octopus · Shell
Claude Code plugin that runs multiple models (Codex, Gemini, Copilot and others) against the same research, design or coding task and surfaces disagreements.
Open sourcekeyword match
darrenhinde/OpenAgentsControl · TypeScript
Plan-first development workflows with approval-based execution, automatic testing, code review and validation, built for OpenCode.
Open sourcekeyword match
Q00/ouroboros · Python
Interview-gated, staged-evaluation, budgeted evolution loop delivered as an MCP server across 14 runtimes including Claude Code, Codex CLI, Gemini CLI and OpenCode.
Open sourcekeyword match
VoltAgent/awesome-codex-subagents · See repository
Ready-to-copy Codex custom subagent .toml files for ~/.codex/agents or .codex/agents, across 13 categories.
Open sourcekeyword match
tintinweb/pi-subagents · TypeScript
Pi extension adding subagents with parallel execution, live widget, fleet view, custom agent types, mid-run steering and Claude-compatible workflows.
Open sourcekeyword match
SethGammon/Citadel · JavaScript
Operating layer for Claude Code and OpenAI Codex: persistent project memory, intent routing, safety hooks, cost telemetry and parallel agent fleets.
Open sourcekeyword match
langchain-ai/langchain-skills · TypeScript
Official LangChain and LangGraph skills for coding agents.
Open sourcekeyword match
NousResearch/hermes-agent-self-evolution · Python
Nous Research layer that optimizes Hermes Agent skills, prompts and code using DSPy and GEPA evolutionary search.
Open sourcekeyword match
Salomondiei08/oh-my-hermes · Shell
oh-my-style workflow layer for Hermes Agent covering building, shipping and operating apps.
Open sourcekeyword match
frankbria/ralph-claude-code · Shell
Runs Claude Code in an autonomous Ralph-style loop with intelligent exit detection and safeguards.
Open sourcekeyword match
stellarlinkco/myclaude · Go
Go-based multi-agent orchestration workflow that coordinates Claude Code with Codex, Gemini and OpenCode.
Open sourcekeyword match
spec-kitty/spec-kitty · Python
Git-native spec-driven workflow: specs define what agents build, a Charter governs how, across coding agents and harnesses.
Open sourcekeyword match
open-gsd/gsd-pi · TypeScript
Community continuation of Get-Shit-Done: meta-prompting and spec-driven development so agents work long autonomous sessions without losing the big picture.
Open sourcekeyword match
modu-ai/moai-adk · Go
Single Go binary that adds SPEC-driven plan/run/sync, TRUST 5 quality gates, model/effort routing and Claude x GLM cost control to Claude Code.
Open sourcekeyword match
vanzan01/claude-code-sub-agent-collective · JavaScript
Research-driven subagent collective using hub-and-spoke coordination and TDD-oriented workflows in Claude Code.
Open sourcekeyword match
donvito/codex-astra-luna-orchestrator · Python
Codex configuration that uses a strong model as orchestrator and a cheaper model for subagents.
Open sourcekeyword match
josstei/maestro-orchestrate · JavaScript
Orchestration layer with 39 specialist agents, parallel subagents, persistent sessions and built-in review/debug/security tooling.
Open sourcekeyword match
nicobailon/pi-subagents · TypeScript
Pi extension for async subagent delegation with truncation, artifacts and session sharing.
Open sourcekeyword match
tiann/hapi · TypeScript
Remote and mobile app for running Codex, Claude Code, Pi, OpenCode, Kimi Code and Grok Build sessions anywhere.
Open sourcekeyword match
Ark0N/Codeman · TypeScript
Self-hosted mission control to run Claude Code, OpenCode, Pi, Codex, Antigravity and Gemini CLI 24/7 from any device and watch subagents live.
Open sourcekeyword match
Palgu/maggy · Python
Maggy ships in one repo with Claude Bootstrap, an installable pack of skills, hooks, rules and templates for Claude Code (it also works with Codex, Kimi and Gemini CLI) that enforces TDD and quality gates and keeps memory across context compaction. Maggy itself is an optional local FastAPI server and web dashboard that adds 13-tier multi-model routing across Claude, DeepSeek, Kimi, Gemini and Codex, intent-driven skill protocols, an MCP server and plugins. It runs in a local mode without API keys. Its routing was later ported to the mixture-of-models Claude Code mod.
Open sourcekeyword match
ZHangZHengEric/Sage · Python
Sage is an MIT-licensed agent harness for developers using Desktop, Server/Studio or the SAgents Python runtime. It supports versioned agent packages, persistent sessions, streamed output, tools, Skills, MCP and multi-agent workflows. Its RSI workflow uses goals or curated history to produce candidate versions with evaluation evidence; users choose whether to adopt them. Desktop RSI currently supports simple types and native tools, not every general MCP integration. Built-in configurations target one process or host; Server supports one worker. Local execution is not container isolation, and persistent state does not provide distributed job takeover or automatic checkpoint resume. AI-assisted submission on behalf of the maintainer, based on the official README; no independent testing is claimed.
Open sourcekeyword match
haabe/mycelium · Python
Your agent can build anything by Sunday. Mycelium makes it ask who it is for first. What it does Mycelium is a plugin for Claude Code that keeps you and your agent honest about whether an idea is worth building, from the first hunch to launch. Before any code: it asks four questions. What is the problem, who has it, what are you assuming, and what would show you wrong. It takes about ten minutes, and the answers go into your repo as the spec. While you build: the agent can't create its first source file until those answers exist. After that it builds only what has a written reason behind it. As you learn: assumptions, evidence and decisions sit in git next to the code. Months later you can tell whether you had a reason or only wanted one. All the way through: the same discipline carries from purpose and strategy to delivery and launch, with checks for security, privacy and accessibility when you get there. Who it's for Solo builders and small teams who build with AI agents and have shipped something nobody used, or would rather not find out the hard way. You know discovery matters. You skip it because the agent is ready now and asking people takes days. What makes it different - It holds the agent, not you. Most product methods are advice you can ignore. Here the agent is stopped, and your side of the trade is ten minutes. - It never calls a guess evidence. What you believe, what one person said and what you have tested are recorded as different things. - It records what you killed. A dropped idea, with the reason, counts as a result. - It is built on named sources. Torres, Christensen, Gilad, Wardley and others, used as working steps and not as slogans. -It is built with itself. Every change to Mycelium goes through Mycelium, and the failures that shaped it are published. - It is free and open source under the MIT licence. It won't make your idea good. It will tell you sooner whether it is.
Open sourcekeyword match
Mintplex-Labs/anything-llm · JavaScript
Local-first application for running a private agent setup, including RAG, multimodal input, computer use, and agent orchestration. It is self-hosted rather than a rented hosted service.
Open sourcekeyword match
Donchitos/Claude-Code-Game-Studios · Shell
Configures Claude Code as a game-development studio with 49 AI agents, 72 workflow skills, and a coordination system mirroring a studio hierarchy. Topics include Claude Code, Unity, Godot, and Unreal Engine.
Open sourcekeyword match
assafelovic/gpt-researcher · Python
GPT Researcher is an autonomous Python agent that conducts deep research on any data using any LLM provider. The Apache-2.0 project includes search, web scraping, and MCP server support.
Open sourcekeyword match
activepieces/activepieces · TypeScript
Activepieces is an open-source TypeScript project for workflow automation with AI agents and MCP servers. The repository describes about 400 MCP servers for AI agents and presents the project as an n8n alternative.
Open sourcekeyword match
czlonkowski/n8n-mcp · TypeScript
An MCP server that lets Claude, ChatGPT, Cursor, and other AI assistants operate n8n. It supports debugging, editing, validating, testing, and building workflows on any n8n instance.
Open sourcekeyword match
1Panel-dev/MaxKB · Python
MaxKB is an open-source platform for building enterprise-grade agents. Project topics include knowledge bases, RAG, LangChain, and an MCP server.
Open sourcekeyword match
cft0808/edict · Python
Python multi-agent orchestration system described as nine specialized AI agents, with a real-time dashboard, model configuration, and full audit trails. MIT licensed.
Open sourcekeyword match
yc-software/qm · TypeScript
qm is a TypeScript project described as a multiplayer agent harness for teams. It is MIT-licensed, and its homepage is on Y Combinator labs. Public topics are ai, assistant, harness, and qm.
Open sourcekeyword match
TencentCloud/Octop · Python
Octop is a self-hosted AI assistant for multi-user, multi-agent use. Repository topics include agentic AI, local-first deployment, and long-term memory.
Open sourcekeyword match
cobusgreyling/loop-engineering · TypeScript
loop-engineering provides patterns, starters, and CLI tools for loop engineering with AI coding agents. Named tools include loop-audit, loop-init, and loop-cost, for systems that prompt and orchestrate agents.
Open sourcekeyword match
EKKOLearnAI/ekko-studio · TypeScript
Ekko Studio is a local-first AI workspace for multi-agent chat, coding, and visual workflows. It is available on desktop and the web.
Open sourcekeyword match
alibaba/spring-ai-alibaba · Java
Spring AI Alibaba is an agentic AI framework for Java developers. Topics cover multi-agent workflows, graphs, ReAct-style agents, and Spring AI.
Open sourcekeyword match
HarnessMD/munder-difflin · JavaScript
Open-source app for running an office of Claude Code, Codex, and similar agents on a laptop, in sandboxes, or elsewhere, using existing subscriptions. It targets local-first multi-agent orchestration, memory, and coding-agent workflows.
Open sourcekeyword match
Orkas-AI/Orkas · JavaScript
Orkas is an MIT-licensed, local-first desktop AI team for macOS and Windows. A Commander turns a goal into to-dos and coordinates 16 built-in specialist agents plus Claude Code, Codex, OpenCode, OpenClaw, and Hermes, with projects, scheduled automations, and connectors. It can use built-in models or your own keys.
Open sourcekeyword match
iflytek/astron-agent · Java
Open-source agentic workflow platform from iFlytek for multi-agent orchestration, MCP, and workflow automation. It includes low-code and workflow-engine components and is released under Apache-2.0.
Open sourcekeyword match
YaoApp/yao · Go
Self-hosted app for keeping agents and workspaces in one place and tracking tasks on a board. It is accessible from desktop, mobile, browser, or API, and is positioned around agent harnesses, orchestration, skills, and coding-agent workflows.
Open sourcekeyword match
Fei2411/DS-Bot · JavaScript
DS-Bot is a JavaScript plugin for DeepSeek Harness that runs teams of long-lived bots. It includes main bots, bot-to-bot messages, and group chats.
Open sourcekeyword match
BuilderIO/agent-native · TypeScript
agent-native is an MIT-licensed TypeScript framework from BuilderIO for building agentic apps. Topics include agent frameworks, MCP, React, and generative UI.
Open sourcekeyword match
xerrors/Yuxi · Python
Yuxi is a self-hosted, multi-tenant knowledge agent platform. It combines RAG, knowledge graphs, multi-agent workflows, MCP and skills, sandboxing, and permission management.
Open sourcekeyword match
loopx-project/loopx · Python
Control plane for always-on agents that turns long-horizon goals into verifiable results across Codex, Claude Code, and custom runtimes. Ranges from local SQLite to team scale. Python, Apache-2.0.
Open sourcekeyword match
agentscope-ai/agentscope-java · Java
AgentScope Java is an open-source framework for building distributed, production-grade, long-running agents. It is written in Java and licensed under Apache-2.0.
Open sourcekeyword match
ModelEngine-Group/nexent · Python
Nexent is an open-source zero-code platform for auto-generating AI agents using harness engineering principles. It unifies tools, skills, memory, and orchestration, with constraints, feedback loops, and control planes. It is written in Python under the MIT license.
Open sourcekeyword match
Waishnav/devspace · TypeScript
TypeScript project described as a minimal coding agent harness on MCP for ChatGPT, Claude, Hermes, Grok Bot, and OpenClaw. Topics include agent orchestration and coding-agent workflows. MIT license.
Open sourcekeyword match
ApodexAI/FrontierAgent · Python
Open-source agent framework with a native command-line TUI, ReAct and Agent Team modes. It is started with one command on macOS and Linux, with no preinstall and no hard Docker dependency.
Open sourcekeyword match
sentient-agi/ROMA · Python
Recursive-Open-Meta-Agent v0.1 (beta), a meta-agent framework for building multi-agent systems. Published by Sentient AGI.
Open sourcekeyword match
SolaceProducts/solace-agent-mesh · Shell
Platform for building and running AI agents across build, test, deploy, monitor, and improve. It hosts agents, runs their tools, and coordinates them over an event mesh, with sessions, artifacts, auth, evaluators, and observability built in.
Open sourcekeyword match
TencentCloudADP/youtu-agent · Python
A Python agent framework for building agents that deliver with open-source models. Repository topics also include OpenAI-style agents.
Open sourcekeyword match
embabel/embabel-agent · Kotlin
Agent framework for the JVM, implemented in Kotlin, for multi-agent systems and orchestration. Apache-2.0 licensed, with Spring among its stated stack.
Open sourcekeyword match
sandbox-quantum/switch · Python
Open-source AI teammates for Slack, Teams, Discord, Telegram, and Mattermost. It is self-hosted and works with any agent harness and any model.
Open sourcekeyword match
kungfu-systems/kungfu · C++
Keeps the same work moving across Codex, Claude, OpenCode, and your own execution surface instead of handing work off between agents. Topics include agent orchestration and agent memory.
Open sourcekeyword match
spacering-net/codeg · Rust
Collaborative coding workspace that aggregates sessions from Claude Code, Codex, OpenCode, Pi, Grok Build, and similar agents. It can run as a desktop app, a self-hosted server, or Docker, and is written in Rust under Apache-2.0.
Open sourcekeyword match
code-yeongyu/lazycodex · TypeScript
Agent harness for complex codebases that covers project memory, planning, execution, and verified completion inside Codex. Repository topics also reference Claude Code and orchestration.
Open sourcekeyword match
tutti-os/tutti · TypeScript
TypeScript desktop app for people and agents to work in a shared, local-first workspace. Topics cover multi-agent orchestration, MCP, and coding agents such as Claude Code and Codex. Apache-2.0.
Open sourcekeyword match
golutra/golutra · Rust
Rust platform that combines Codex, Claude Code, and OpenClaw into one agent system with parallel execution, task orchestration, and long-running workflows. It is aimed at automation and developer workflows, and ships as a desktop app.
Open sourcekeyword match
NVlabs/SoL-Pi · TypeScript
NVlabs project on scaling auto-research loops for efficient agent harnesses. TypeScript codebase under the MIT license, with a project page.
Open sourcekeyword match
SeemSeam/claude_codex_bridge · Python
Python CLI workspace for running Codex, Claude, Gemini, Kimi, Qwen, Cursor, Copilot, Pi, OpenCode, and other coding agents together in one visible multi-agent session.
Open sourcekeyword match
michael-denyer/pstack-claude · TypeScript
Claude Code, Codex, Copilot, Pi, OpenCode, Muse, Gemini, and Prime Agent versions of Poteto's pstack. Translates Cursor workflow primitives into agent workflows for those harnesses.
Open sourcekeyword match
open-gitagent/opengap · TypeScript
opengap is a framework-agnostic, git-native standard for defining AI agents. Topics include agent skills and an open standard rather than a single runtime.
Open sourcekeyword match
MiroMindAI/MiroFlow · Python
MiroFlow is a Python agent framework that includes a web UI and supports MiroThinker, Claude, Kimi, and OpenAI. Repository topics center on research agents and agent benchmarks.
Open sourcekeyword match
SciSharp/BotSharp · C#
BotSharp is an AI multi-agent framework in .NET, written in C#. Its listed focus is chatbots and multi-agent systems.
Open sourcekeyword match
teaql/teaql-agent-kit · TypeScript
A model-driven harness for AI coding agents. It covers semantic modeling, app generation, progressive API assist, and evidence-based verification.
Open sourcekeyword match
CoderLuii/HolyClaude · JavaScript
An AI coding workstation that combines Claude Code, a web UI, eight AI CLIs, a headless browser, and more than 50 tools.
Open sourcekeyword match
amElnagdy/delegate-skills · JavaScript
Skills for delegating a coding task to a separate coding agent CLI, reviewing the diff, and landing the commit yourself. They are split one per implementer and target CLIs such as Claude Code, Codex, Cursor, and OpenCode.
Open sourcekeyword match
777genius/agent-teams-ai · TypeScript
TypeScript app where agents take tasks, message each other, and review work while the user directs them from a kanban board. It names support for Codex, Claude, OpenCode, Cursor, and other coding-agent runtimes.
Open sourcekeyword match
agentuniverse-ai/agentUniverse · Python
agentUniverse is an Apache-2.0 Python framework for building LLM multi-agent applications. The repo is tagged for agents, autonomous systems, and multi-agent workflows.
Open sourcekeyword match
skalesapp/skales · Unknown
Skales is a personal AI agent harness for macOS, Windows, Linux, Android, and iOS, with chat, coding (Skales Code), Flow, automations, computer use, browser, voice, MCP, BYOK across 26 providers, and local LLMs via Ollama. Skales Scout runs lasting desk agents, optionally each in its own virtual machine. It does not use Docker or a terminal.
Open sourcekeyword match
unreallabsai/unreal-agent · Go
unreal-agent is an async-first agent harness written in Go and released under the MIT license.
Open sourcekeyword match
GammaLabTechnologies/harmonist · Python
Harmonist is a Python project for portable AI agent orchestration with mechanical protocol enforcement. Its description cites 186 agents and zero runtime dependencies, and topics include multi-agent frameworks, Claude Code, and Cursor.
Open sourcekeyword match
neuron-core/neuron-ai · PHP
Neuron AI is an agentic framework for the PHP ecosystem for building AI-driven applications. It connects LLMs, tools, vector databases, and memory to agents that interact with application data and UI.
Open sourcekeyword match
zhnt/loushang · Python
Loushang is a Python agent harness for coding workflows. It covers multi-model LLM orchestration, stateful sessions, tool governance, traceable delivery, and provider routing for GPT, Claude, DeepSeek, Qwen, Kimi, GLM, and MiniMax.
Open sourcekeyword match
trpc-group/trpc-agent-go · Go
A Go framework for building production agent systems. It includes graph workflows, tools, memory, A2A, AG-UI, MCP, evaluation, and observability.
Open sourcekeyword match
agentjido/jido · Elixir
An autonomous agent framework for Elixir, built for distributed autonomous behavior and dynamic workflows.
Open sourcekeyword match
Roy3838/Observer · TypeScript
Observer is an open-source TypeScript micro-agent orchestrator that runs completely locally and monitors things for you. It is licensed under AGPL-3.0.
Open sourcekeyword match
Core-Mate/OpenGUI · HTML
OpenGUI is an Android GUI agent framework for phone-use AI that can see, plan, and operate real mobile apps through the GUI.
Open sourcekeyword match
exoharness/exo · Rust
Exo is an agent and harness architecture that is fully recursive and can edit all aspects of itself at runtime to get better at tasks. It is written in Rust and licensed under MIT.
Open sourcekeyword match
AgentEra/Agently · Python
Agently is a Python framework for building GenAI applications and interacting with agents in code through structured data and chained calls. It uses an event-driven flow, TriggerFlow, for working logic and can switch models without rewriting application code. It is licensed under Apache-2.0.
Open sourcekeyword match
codeaholicguy/ai-devkit · TypeScript
TypeScript project described as a control plane for AI coding agents. Repository topics include Claude Code, Codex CLI, Cursor, Gemini CLI, OpenCode, MCP, and memory. Licensed Apache-2.0.
Open sourcekeyword match
langchain-ai/deepagentsjs · TypeScript
TypeScript project from langchain-ai described as a batteries-included agent harness. It is MIT-licensed and documented under the LangChain JavaScript Deep Agents overview.
Open sourcekeyword match
agentlas-ai/Agentlas-OS · Python
Python agent OS that keeps specialist agents in a hub and spins up a temporary orchestrator per task. It is local-first, works with any model, and lists MCP and multi-agent topics. Apache-2.0.
Open sourcekeyword match
autonomous-ai/openharness · TypeScript
OpenHarness is an open-source TypeScript harness for managing multiple coding agents across machines from one command center. Its topics include Claude Code, Codex, OpenCode, tmux and MCP, and it extends to areas such as CAD, circuits and robotics.
Open sourcekeyword match
inkeep/agents · TypeScript
Inkeep Agents is a self-hostable platform for creating AI agents and multi-agent workflows, using either a no-code visual builder or a TypeScript SDK that stay in sync. It includes MCP support and developer SDK tooling.
Open sourcekeyword match
ThousandBirdsInc/chidori · Rust
Chidori is a Rust agent framework where runs are durable, replayable, and resumable by default. It includes checkpointing, deterministic execution, and replay, plus a JavaScript engine with TypeScript support.
Open sourcekeyword match
thClaws/thClaws · Rust
thClaws is an open-source AI agent harness written in Rust. One binary provides GUI, CLI, headless, and webapp interfaces, with multi-provider support, MCP, skills, plugins, and agent teams.
Open sourcekeyword match
Chorus-AIDLC/Chorus · TypeScript
Chorus is an agent harness for AI-human collaboration, inspired by the AI-Driven Development Lifecycle. The TypeScript project covers multi-agent work, Claude Code, and kanban-style task management, and is licensed AGPL-3.0.
Open sourcekeyword match
polyuiislab/infiAgent · Python
A Python agent framework for long-running, complex tasks. Agents are set up by editing config files, and it supports Anthropic skills. Licensed GPL-3.0.
Open sourcekeyword match
muratcankoylan/Agent-Skills-for-Context-Engineering · Python
Collection of agent skills for context engineering, multi-agent architectures, and production agent systems. Intended for building, optimizing, or debugging agent systems that depend on effective context management.
Open sourcekeyword match
refly-ai/refly · TypeScript
Refly lets you define skills through a workflow approach and run them on Claude Code, Cursor, Codex and other agents. It can also expose them as APIs or Slack and Lark/Feishu bots.
Open sourcekeyword match
breaking-brake/cc-wf-studio · TypeScript
VS Code extension for Claude Code workflow automation, tagged for agent skills, MCP tools, slash commands, and sub-agents. The extension is listed on Open VSX.
Open sourcekeyword match
rpamis/comet · JavaScript
Comet is an agent skill harness for turning ideas into evaluated workflows. It focuses on skill creation, specs, phase-guarded loops, and evals.
Open sourcekeyword match
a5c-ai/babysitter · JavaScript
Babysitter is a plugin for Claude Code, Codex, Hermes, and Pi that orchestrates agent work through deterministic workflow steps. It targets complex, multi-step agent tasks rather than one-off prompts.
Open sourcekeyword match
AMAP-ML/LongHorizon-Harness · Python
A computer-use harness for running AI agents across desktop apps and the CLI for extended periods while preserving task state. It provides fresh-context execution, durable verified state, independent auditing, and recoverable progress, with native Claude Code, Codex, and OpenClaw integration.
Open sourcekeyword match
hesamsheikh/octogent · TypeScript
A thin orchestration dashboard over Claude Code for managing context, automation, and developer headspace. It targets the Claude Code and Codex harnesses.
Open sourcekeyword match
agentrq/agentrq · Go
Self-hosted realtime conversational task manager for controlling AI agents from mobile, web, or desktop. It is designed to work with your own Claude subscriptions and any harness with ACP support.
Open sourcekeyword match
keli-wen/agy-staff · JavaScript
agy-staff is a plugin that uses Google's Antigravity CLI as a Gemini staffer for Claude Code, OpenAI Codex, and DeepSeek Harness. It is listed for Claude Code, Codex, OpenCode, Gemini CLI, and Antigravity.
Open sourcekeyword match
sno-ai/sno-station-skills · Shell
Open-source skills for Claude Code, Codex, Cursor, and Hermes Agent. They cover review by a different agent, a handover before a usage limit stops the work, treating done as proven by a real run, and a nightly loop that learns from your own sessions.
Open sourcekeyword match
kingbootoshi/cartographer · TypeScript
A TypeScript Claude Code plugin that maps and documents codebases using parallel AI subagents.
Open sourcekeyword match
23blocks-OS/ai-maestro · TypeScript
An AI agent orchestrator with a skills system for managing Claude Code, Codex, Grok Build, or other agents from one dashboard. It provides memory search, code graph queries, and agent-to-agent messaging, and can move agents between computers and locations.
Open sourcekeyword match
hamelsmu/claude-review-loop · Shell
Claude Code plugin that runs an automated code review loop with Codex. It is listed for both the Claude Code and Codex harnesses.
Open sourcekeyword match
gmickel/flow-next · Python
Workflow plugin that takes a bug, idea, or ticket to a verified pull request, with specs, cross-model review by risk, live QA, and receipts in the repo. It lists Claude Code, Codex, Cursor, Droid, Grok Build, OpenCode, and Muse Code.
Open sourcekeyword match
shinpr/claude-code-workflows · JavaScript
Development workflows for Claude Code that keep broad exploration focused on an outcome you approved. Repository topics include agent skills, subagents, code review, and LLM orchestration.
Open sourcekeyword match
aannoo/hcom · Rust
Lets AI agents message, watch, and spawn each other across terminals. Supports Claude Code, Codex, Antigravity CLI, Cursor CLI, OpenCode, Kilo, Pi, and Kimi.
Open sourcekeyword match
Hainrixz/the-architect · Unknown
Claude Code plugin that interviews you, designs the architecture, and writes a self-contained blueprint another Claude Code instance builds from. Each build step includes EARS acceptance criteria and a runnable verify command. It covers 14 project shapes, greenfield and brownfield, in English and Spanish.
Open sourcekeyword match
tzachbon/smart-ralph · Shell
Claude Code plugin for spec-driven development that combines a Ralph Wiggum loop with a structured specification workflow and smart compaction. Repo topics also list Codex skills support.
Open sourcekeyword match
zscole/adversarial-spec · Python
A Claude Code plugin that iteratively refines product specifications by debating them across multiple LLMs until the models reach consensus.
Open sourcekeyword match
win4r/openclaw-a2a-gateway · TypeScript
OpenClaw plugin that implements Agent-to-Agent (A2A) protocol v0.3.0 as a bidirectional communication gateway between agents.
Open sourcekeyword match
ZaxbyHub/opencode-swarm · TypeScript
OpenCode plugin for architect-centric multi-agent orchestration. It uses hub-and-spoke coordination for specialist consultation, code generation, and QA review.
Open sourcekeyword match
prevalentWare/opencode-goal-plugin · TypeScript
OpenCode plugin that adds Codex-style goal mode, /goal slash commands, and persistent objectives for a coding agent.
Open sourcekeyword match
yuting0624/antigravity-for-claude-code · Shell
Community Claude Code plugin that runs the Antigravity CLI (Gemini) as a collaborating sub-agent, with model routing across the software development lifecycle. It is not affiliated with Google or Anthropic. Shell, MIT license.
Open sourcekeyword match
boshu2/agentops · Go
DevOps discipline for AI coding agents that shapes the work, tracks it as a graph, and has each change judged by a context that did not write it. Plugins target Claude Code, Codex, OpenCode, and Cursor.
Open sourcekeyword match
prekuter/dryforge · Python
Dryforge is a bounded-autonomy plugin harness for agents, described as taking work from intent to implementation. It is a Python plugin for Claude Code, Codex, GitHub Copilot, and Antigravity.
Open sourcekeyword match
tobihagemann/turbo · Shell
turbo is an agentic software factory for Claude Code and Codex, packaged as plugins and agent skills. It is for shipping code changes of any size.
Open sourcekeyword match
spencermarx/open-code-review · TypeScript
AI-powered multi-agent code review for Claude Code and OpenCode. It simulates a customizable team of engineers performing code review, with built-in discourse.
Open sourcekeyword match
Ibrahim-3d/orchestrator-supaconductor · Python
Multi-agent orchestration system for Claude Code with parallel execution, automated quality gates, a Board of Directors, and bundled Superpowers skills.
Open sourcekeyword match
CronusL-1141/AI-company · Python
A Claude Code mod for multi-agent teams. It describes MCP tools, agent templates, lifecycle hooks, pipeline workflows, persistent teams, structured meetings, a task wall, and a React dashboard, with native Claude Code integration rather than LangChain or AutoGen.
Open sourcekeyword match
huisezhiyin/sdd-riper · Python
Lightweight AI agent harness for agentic coding. Strong models explore while humans steer with minimal specs, checkpoints, approval, validation, and reverse sync. It targets Claude Code, Cursor, and OpenClaw.
Open sourcekeyword match
mdiri-ai/SoduWork · JavaScript
SoduWork is an open-source, self-hosted engine for teams and their agents. It is described as multi-harness and multi-agent, with topics covering workflows, MCP, a plugin system, human-in-the-loop collaboration, and harnesses such as Claude Code, Codex, and OpenCode.
Open sourcekeyword match
Tencent/LoopForge · Go
LoopForge is a self-hosted workflow system for coding-agent delivery. It orchestrates CodeBuddy, Codex, and Claude Code with scripts and human approval gates, and includes local workflows for coding assistants.
Gatheringkeyword match
Nov 14, 2026 · Online
A live online workshop on harness engineering: the design decisions around the model, what code should enforce, and approval gates. Covers the internals of Claude Code, DeerFlow and MiroFlow (10:00–13:30 ET).
Gatheringkeyword match
Nov 17, 2026 · Austin, TX
An Agentics Foundation meetup in Austin (18:00–20:00 CT) with rUv, creator of the open-source tools Ruflo and RuView, on agent orchestration, persistent memory and self-hosted agent infrastructure.
Gatheringkeyword match
Dec 19, 2026 · Online
A three-hour online workshop (09:30–12:30 ET) building financial research agents in LangGraph on real SEC filings, using agentic patterns for planning, evidence checking, memory management and cost control.
Gatheringkeyword match
Dec 15, 2026 · Online
A full-day online workshop (10:00–17:00 CET) building agents three ways: zero-dependency Java, Quarkus/MicroProfile with LangChain4j, and serverless on AWS Bedrock AgentCore. Also covers memory architectures, subagents, skills and eval loops.