Browse agents, frameworks and supporting tools. Use categories to narrow the task, then save candidates to compare.
Superagent for Mac is an open-source macOS workspace for Claude Code and Codex, combining persistent coding-agent chats with an agent-controlled browser, iOS Simulator, and companion iPhone app for monitoring conversations and responding to agent requests. It runs locally and uses the user’s existing agent subscription.
SVRF is a merge train that batches, gates and lands pull requests for repositories where coding agents open pull requests faster than CI can gate them one by one; it ships an MCP server (svrf mcp) so an agent whose pull request is held can query why and fix it itself.
Swarms orchestrates multi-agent collaboration for production-grade applications, solving issues like short memory and high costs, with customizable tools for specific needs, currently used by RBC, John Deere, and AI startups
This open source project introduces SWE-agent, a software engineering agent for LMs like GPT-4, enhancing bug and issue resolution in GitHub repositories with state-of-the-art performance, facilitated by a well-designed Agent-Computer Interface (ACI) and support for OpenAI and Anthropic Claude models
A local-first, self-hosted agent-to-agent communication platform with natural-language messaging, human-approved tasks, groups, delegation, a CLI, Python client, web UI, and encrypted backups.
Tabby is a self-hosted, open-source AI coding assistant similar to GitHub Copilot, featuring a self-contained setup with no DBMS/cloud dependency, OpenAPI for easy integration, consumer-grade GPU support, and a full-feature admin UI in its latest release
Tale is a collaborative workspace where teams turn work into tasks, delegate tasks to configured agents in persistent sandbox workspaces, and review their reports and deliverables.
An AI-native workspace with custom agents, workflow automation, and prompt-built apps; its official public repository contains product documentation and clone-ready App Kits.
An open-source security and governance SDK for AI applications, providing deterministic policy enforcement, guardrails, cost tracking, and audit evidence.
A self-hosted workspace for coordinating AI agent teams with structured tasks, persistent memory, and file-conflict controls.
The Agentic Leaderboard publishes weekly rankings of open-source AI agents across reliability, tool selection, autonomous iteration, and community signals, with public data and methodology.
The Forge coordinates LLMs to generate code candidates, compiles and benchmarks them, then evolves top candidates through mutation and selection.
An open standard that defines eight evidence-based dimensions for AI agents assessing human–AI operator interactions and issuing credentials.
A portable set of standards for AI coding agents covering repository discovery, language selection, development workflows, and post-generation audits.
Thursday is an open-source personal voice assistant that runs on your computer and delegates longer tasks to background AI agents using a real browser, shell, files, Agent Skills, and MCP servers. It keeps the live call available while work runs and returns results as local files; it uses hosted model providers rather than local inference.
A local-first AI agent framework for running specialist agents with shared memory, tools, approvals, and mission control.
Tree Ring Memory is a framework-agnostic, local-first memory lifecycle for AI agents, with SQLite search, deliberate forgetting, audit logs, consolidation, and runtime adapters.
TypeChat is a library that facilitates building natural language interfaces by using schema engineering as an alternative to traditional function calling in LLMs, avoiding JSON schema-based constraints
