Related Agent Harnesses¶
🚀 Trending (fastest growth)¶
| # | Project | Growth | Stars | Description |
|---|---|---|---|---|
| 1 | Cordis | ⬆️ +266 | ⭐6,867 | Meta-Framework of Spatiotemporal Composability — the plugin runtime DeepSeek Harness is built on. |
| 2 | DeerFlow | ⬆️ +55 | ⭐80,462 | Open-source long-horizon SuperAgent harness by ByteDance: skills, memory, sandboxes, subagents, tools and a message gateway. |
| 3 | agentmemory | ⬆️ +36 | ⭐27,233 | Persistent memory for AI coding agents based on real-world benchmarks (upstream of DSH agentmemory ports). |
| 4 | mnemon | ⬆️ +6 | ⭐500 | LLM-supervised persistent memory for AI agents: graph-based recall and cross-session knowledge in a single binary. |
| 5 | deeptide | ⬆️ +3 | ⭐1,091 | Swift-native macOS coding agent built by DeepSeek, for DeepSeek. |
🔥 Top 10¶
| # | Project | Stars | Description | Status |
|---|---|---|---|---|
| 1 | DeerFlow | ⭐80,462 | Open-source long-horizon SuperAgent harness by ByteDance: skills, memory, sandboxes, subagents, tools and a message gateway. | ✅ active |
| 2 | ruflo | ⭐69,896 | The original agent meta-harness: deploy intelligent multi-player swarms, coordinate autonomous workflows, adaptive memory, self-learning intelligence, RAG integration and native Claude Code / Codex integrations. | ✅ active |
| 3 | CodeWhale | ⭐40,830 | Open-source, community-driven agent harness. | ✅ active |
| 4 | DeepSeek-Reasonix | ⭐35,284 | DeepSeek-native AI coding agent for your terminal, engineered around prefix-cache stability — leave it running. | ✅ active |
| 5 | agentmemory | ⭐27,233 | Persistent memory for AI coding agents based on real-world benchmarks (upstream of DSH agentmemory ports). | ✅ active |
| 6 | Cordis | ⭐6,867 | Meta-Framework of Spatiotemporal Composability — the plugin runtime DeepSeek Harness is built on. | ✅ active |
| 7 | deeptide | ⭐1,091 | Swift-native macOS coding agent built by DeepSeek, for DeepSeek. | ✅ active |
| 8 | sandbase-harness | ⭐628 | Open-source CMA-compatible agent runtime for any model: MCP tools, sandboxed sessions, audit, replay. | ✅ active |
| 9 | mnemon | ⭐500 | LLM-supervised persistent memory for AI agents: graph-based recall and cross-session knowledge in a single binary. | ✅ active |
| 10 | claude-paper | ⭐324 | Cross-agent research paper toolkit for Claude Code, Codex, OpenCode and DeepSeek Harness: quick summaries and deep dives. | ✅ active |
Complete list (13)¶
Harness (13)
| Project | Stars | Description | Status |
|---|---|---|---|
| DeerFlow | ⭐80,462 | Open-source long-horizon SuperAgent harness by ByteDance: skills, memory, sandboxes, subagents, tools and a message gateway. | ✅ active |
| ruflo | ⭐69,896 | The original agent meta-harness: deploy intelligent multi-player swarms, coordinate autonomous workflows, adaptive memory, self-learning intelligence, RAG integration and native Claude Code / Codex integrations. | ✅ active |
| CodeWhale | ⭐40,830 | Open-source, community-driven agent harness. | ✅ active |
| DeepSeek-Reasonix | ⭐35,284 | DeepSeek-native AI coding agent for your terminal, engineered around prefix-cache stability — leave it running. | ✅ active |
| agentmemory | ⭐27,233 | Persistent memory for AI coding agents based on real-world benchmarks (upstream of DSH agentmemory ports). | ✅ active |
| Cordis | ⭐6,867 | Meta-Framework of Spatiotemporal Composability — the plugin runtime DeepSeek Harness is built on. | ✅ active |
| deeptide | ⭐1,091 | Swift-native macOS coding agent built by DeepSeek, for DeepSeek. | ✅ active |
| sandbase-harness | ⭐628 | Open-source CMA-compatible agent runtime for any model: MCP tools, sandboxed sessions, audit, replay. | ✅ active |
| mnemon | ⭐500 | LLM-supervised persistent memory for AI agents: graph-based recall and cross-session knowledge in a single binary. | ✅ active |
| claude-paper | ⭐324 | Cross-agent research paper toolkit for Claude Code, Codex, OpenCode and DeepSeek Harness: quick summaries and deep dives. | ✅ active |
| open-managed-agents | ⭐243 | Open-source Claude Managed Agents API implementation and self-hosted Claude Tag-style agent runtime. | ✅ active |
| Axern | ⭐57 | Open-source sandboxes for AI agents: untrusted code execution and durable services. | ✅ active |
| deepseek-auto-evolving-harness | ⭐28 | Auto-evolving LLM agent harness: benchmark-driven evolution via Claude Code and a self_evolution.md guide. | ✅ active |