Agent Frameworks

Open-source desktop AI coworker for browsing and code execution.

Accomplish Coworker is an MIT-licensed open-source AI coworker that runs on the desktop, combining computer-use style browsing with code execution so agents can research, implement, run, and debug workflows in one local environment.

Context compression for LLM apps and coding agents

Headroom is an Apache-2.0 context compression layer for LLM apps and coding agents. It compresses tool output, logs, files, RAG chunks, and agent history through a local library, proxy, wrapper, or MCP server, with retrieval hooks for bringing originals back when needed. Treat its savings numbers as Headroom-reported benchmarks, not independent aicoolies measurements.

Codebase knowledge graph MCP server for AI coding agents

Codebase Memory MCP is an MIT-licensed MCP server that turns a repository into a persistent code knowledge graph for AI coding agents. It gives Claude Code, Cursor, Codex-style agents, and other MCP clients structural queries for functions, classes, call chains, routes, and architecture, helping them explore large projects without repeatedly rereading files or relying only on broad search.

Python and TypeScript framework for production multi-agent systems

BeeAI Framework is an Apache-2.0 toolkit for building production-ready AI agents and multi-agent systems in Python and TypeScript. Its docs cover agents, tools, RAG, memory, workflows, backend providers, serving, and A2A/MCP integration surfaces, making it a vendor-neutral option for teams comparing LangGraph, CrewAI, Mastra, and related agent runtimes.

MCP integration platform for agent tool use at scale

Klavis AI is an Apache-2.0 MCP integration platform for teams connecting AI agents to external SaaS tools and APIs. The public repo and official docs position it as infrastructure for reliable tool access at scale, so it fits teams that want reusable MCP connectors without treating every integration as a one-off script or custom OAuth maintenance project.

Open-source Firecracker sandboxes for long-running AI agents

Superserve is an open-source sandbox infrastructure layer for AI agents that need durable computers instead of short-lived shells. It runs isolated Firecracker microVMs, supports pause, resume, snapshot, fork, preview URLs, MCP connectivity, SDK/API control, Docker workloads, and self-hosting, while the hosted service adds pay-as-you-go agent sandboxes for teams.

MCP gateway and integration catalog for AI agents

Executor is an MIT-licensed integration layer and MCP gateway for AI agents. It gives Claude Code, Cursor, Codex, and other MCP-speaking clients one endpoint for connected OpenAPI specs, GraphQL APIs, MCP servers, Google Discovery sources, and custom JavaScript tools, with local, cloud, and self-hosted deployment options for teams centralizing tool access.

Sentry-style observability for AI agent conversations

Latitude is an agent observability platform for teams that need to inspect LLM traces, conversations, issues, and evaluation feedback in one workflow. Its public repo and docs position it as a Sentry-style monitor for AI agents, with semantic search, issue detection, annotations, MCP-assisted fixes, and cloud or self-hosted deployment paths for production debugging.

Filesystem-first framework for durable AI agents

Eve is Vercel's filesystem-first TypeScript framework for building durable AI agents as ordinary project files. It combines Markdown instructions and skills, typed tools, channels, connections, subagents, schedules, sandboxes, and evals with Vercel's agent runtime so teams can ship deployable agents without hand-rolling orchestration. The current beta fits Vercel-native backend agent projects.

Open-source agentic browser that runs local AI agents in your browsing workflow.

BrowserOS is a privacy-first, open-source agentic browser for running AI assistants locally inside real browsing sessions instead of handing every task to a remote cloud browser.

Microsoft’s public-preview runtime governance toolkit for policy, identity, sandboxing, audit, and MCP security around AI agents.

Agent Governance Toolkit is Microsoft’s MIT-licensed public-preview toolkit for governing AI agent runtimes. It adds policy enforcement, zero-trust identity, execution sandboxing, audit, reliability, and MCP security-gateway patterns around tool calls and autonomous actions, helping platform teams move beyond prompt-only guardrails while preserving architecture review requirements.

Microsoft’s pytest-native red teaming framework for turning AI agent safety findings into CI tests.

RAMPART is an open-source Microsoft framework for safety and security testing of agentic AI applications. It brings red-team findings into a pytest-native workflow so teams can turn prompt injection, unsafe tool use, and behavioral boundary failures into repeatable regression tests. The strongest aicoolies angle is developer workflow: RAMPART makes agent safety part of CI/CD instead of a one-off security review.

Local-first personal AI agent with memory trees, desktop integrations, and private workspace context.

OpenHuman is an open-source, local-first personal AI agent from TinyHumans. It combines a desktop app, persistent memory trees, Obsidian-compatible storage, OAuth integrations, and local model support into a private assistant harness. It is most interesting for users who want agentic workflows and long-term memory without handing every context detail to a fully cloud-hosted assistant.

MCP-native personal context vault for keeping AI agents aligned with your work, voice, and projects.

Unabyss is a personal context headquarters for AI agents. It syncs sources such as email, Slack, Notion, Drive, meetings, and professional profiles into structured context files that can be served to MCP-capable clients. The strongest angle is not generic note taking; it is permissioned, reusable context for Claude, Cursor, custom agents, and other tools that otherwise need the same background explained repeatedly.

Version control for AI coding-agent actions

Re_gent is an open-source version-control layer for AI coding-agent activity. Instead of only reviewing the final Git diff, it records what the agent attempted, changed, and executed along the way so teams can trace, undo, and govern autonomous coding work. It fits Claude Code, Codex, Cursor, and multi-agent teams that need an audit trail between prompt and pull request.

Persistent memory layer for AI coding agents — keeps Claude Code, Codex, Cursor, and any MCP agent in context across sessions

agentmemory is an open-source MCP server that gives AI coding agents persistent, cross-session memory. Built on hybrid vector-graph search, it achieves 95.2% recall on the LongMemEval-S benchmark while using up to 92% fewer context tokens than naive context injection. Works out of the box with Claude Code, Codex, Cursor, Windsurf, Cline, OpenCode, Kilo Code, Hermes, and any MCP client through 51 MCP tools plus 12 hooks and 4 skills.

MCP, ACP and Skills support for building production coding agents — interactive or automated.

fast-agent is an Apache-licensed Python framework for building and running LLM agents with full MCP (Model Context Protocol) and ACP support. It ships with an interactive shell mode, Skills management, and multi-model routing — making it a practical platform for coding agents, workflow automation, and agent evaluation across Claude, Codex, HuggingFace, and local models.

Command center for Claude Code and Codex — monitor, steer, and voice-control your AI agents from any device.

Omnara is a command center for Claude Code and Codex sessions across desktop, web, mobile, and Apple Watch. Its public site supports cross-device supervision, parallel agents, worktrees, Git, and session-continuity language, with free-offer metadata; teams should verify detailed pricing, relay, privacy, and enterprise controls before standardizing it.

Vectorless, reasoning-based RAG that reads documents like a human expert — no vector DB, no chunking.

PageIndex is a vectorless, reasoning-based RAG system that builds hierarchical tree indexes from long documents and uses LLMs to navigate them like a human expert would. Instead of chunking text and comparing embeddings, it constructs a table-of-contents-style structure and reasons its way to the right sections — no vector database required. Available as an open-source Python package, cloud API, MCP server, and chat platform.

Production-grade browser automation with AI self-healing and Playwright code ownership

Intuned is a code-first browser automation platform that turns natural language prompts into production-ready Playwright code, deploys it, and self-heals it when target sites change. Supports TypeScript and Python with Anthropic Computer Use, OpenAI CUA, Stagehand, Browser-Use, and Gemini Computer Use integrations. Built-in stealth, captcha solving, auth session management, and scheduled runs with concurrency control. No vendor lock-in—you own the code.

Open-source post-building layer for agents — tracing, evals, and online monitoring

Judgeval is the open-source post-building layer for AI agents from Judgment Labs, providing OpenTelemetry-based tracing, hosted and custom evaluation scorers, and online behavior monitoring for LLM-powered applications. Instrument any function with a single decorator, score live production traffic against faithfulness and instruction-adherence checks, and feed real-world failures back into reinforcement learning or supervised fine-tuning loops.

Open-source observability and self-healing layer for AI agents

TraceRoot is a YC S25-backed open-source observability platform purpose-built for AI agents and LLM apps. It combines OpenTelemetry-compatible tracing with an agentic debugging runtime that reads your source code, correlates failures with recent commits, and proposes fix PRs automatically. BYOK support spans seven LLM providers; the entire stack runs self-hosted via Docker Compose, with TraceRoot Cloud available for managed deployments.

Always-on cloud engineer that lives in Slack and ships verified PRs

Roomote is a Slack-first cloud coding agent from RooCodeInc that takes prompts end-to-end across GitHub, Linear, Notion, Sentry, and your own dev environment, then opens self-verified pull requests for review. It is the team behind 24K+ star Roo Code going all-in on cloud agents: mention it in Slack, let it run in an isolated environment, and review the PR through your normal GitHub flow.

Sandboxes for coding agents — Linux VMs, Git, and deploys in one box

Freestyle is YC-backed sandbox infrastructure built for AI coding agents, shipping secure Linux VMs with nested virtualization, Git servers, and one-click web deploys. It lets agents run real workloads, branch repos, and deploy apps under short-lived identities while billing only for active compute. Used in production by vly.ai, Rork, and Vibeflow.