Open Source

The README opens by calling itself an off-the-shelf agent harness: MIT licensed, Python (plus deepagents.js), built on LangGraph, so streaming, durability and checkpoints are inherited rather than rewritten. The key to reading it is the acknowledgement - inspired by Claude Code, trying to figure out what makes it general and pushing that further - which makes this a reverse-engineering exercise: take a closed coding agent that is known to work well, open it up, and ask what actually lets it handle long tasks. Its answer is four bundled things: a pluggable-backend file system (the artefacts of long tasks are files, not messages, so context keeps only pointers), sub-agents each with their own window (context isolation and compression, not extra helpers), long-thread summarisation with tool output offloaded to disk, and on-demand Skills. Also sandboxed shell execution, cross-session persistent memory, human-in-the-loop approval/edit/reject before a tool call runs, and any MCP server as a tool. Model agnostic: frontier APIs, open weights hosted on Baseten or Fireworks, and self-hosted Ollama/vLLM/llama.cpp all work, with three lines of create_deep_agent giving you a planning, file-reading-and-writing agent. The security section is unusually blunt - it follows a trust-the-LLM model, boundaries must be enforced at the tool and sandbox layer, and you should not expect the model to restrain itself. 29.7k stars. Not run yet; sub-agent isolation quality and summarisation information loss are unverified, so this is graded as pending reproduction.

Agent HarnessCoding AgentSub-Agents

Python30k4.2k139

ahujasid/mcp-for-blenderMIT

MCP for Blender: 32 tools that make Blender a 3D workspace an LLM can drive

Formerly blender-mcp, a third-party community plugin (the README states on its first screen that it is not made by Blender), MIT licensed, Python, and the highest-starred entry among 3D MCP servers. The PyPI package is now mcp-for-blender, yet uvx blender-mcp still runs and existing configs need no change. Two components only: an addon inside the Blender process runs a socket server that actually executes commands (the only place with bpy, the current scene and the UI thread), while a separate MCP server process translates protocol, joined by JSON over TCP on localhost:9876 by default. The source defines 32 @mcp.tool() decorators across four layers - perception (get_scene_info, get_object_info, get_viewport_screenshot returning an MCP Image, closing an act-then-look loop), reference lookup (bpy_api_lookup and describe_node_type so the model queries instead of guessing socket order and enum names), execution (execute_blender_code as the escape hatch, set_texture, export_scene to GLB/FBX), and asset retrieval and generation (Poly Haven with semantic search, importing the .blend source and writing attribution properties; Poly Pizza with about 10,600 low-poly models, size normalisation and automatic CC-BY credit; Sketchfab; Hyper3D Rodin text-to-3D and image-to-3D; Tencent Hunyuan3D distinguishing the mainland ai3d endpoint from the international hunyuan one). On security, arbitrary code execution is the design rather than a bug: an optional safe mode (BLENDER_MCP_SAFE_MODE=1) statically validates scripts before running, blocking file access, subprocesses, network and persistent code while modelling, rendering, saving, import and export pass normally, and blocked scripts return to the model with the reason so it can retry. Content telemetry is off by default, and a dedicated disable_telemetry tool can only turn collection off. 29.2k stars. Not run with Blender locally yet, so this is graded as pending reproduction.

MCPBlender3D Generation

Python29k2.7k202

GLips/Figma-Context-MCPMIT

Framelink MCP for Figma: design files compressed into usable context, not screenshots

Also known as Figma-Context-MCP, npm package figma-developer-mcp, TypeScript, MIT, and a third-party implementation rather than the Dev Mode MCP server Figma ships itself. It exposes only two tools: get_figma_data (read-only) and download_figma_images (removable with –skip-image-downloads, and confined to writing inside –image-dir). The narrow surface is deliberate; the engineering sits in the fetch → simplify → serialize pipeline behind it. Simplify uses composable extractors (layout, text, visuals, component, plus the allExtractors, layoutAndText, contentOnly, visualsOnly and layoutOnly bundles) to compress the raw Figma response into semantic layout and styling in a single tree walk, with maxDepth and nodeFilter, and outputs tree (default, cheapest in tokens), yaml or json. The part worth copying is that compression effectiveness is a metric computed on every call: rawSizeKb against simplifiedSizeKb, rawNodeCount against simplifiedNodeCount, component/instance/text/image node counts, namedStyleCount (the source comment states a high count is a design-system maturity signal), hasVariables (whether Figma Variables are used at all), and separate fetch/simplify/serialize timings; the same hooks drive MCP progress notifications and heartbeats so a large file taking tens of seconds does not look hung. Judgement shows in the details: node IDs accept both the plain and the deeply nested instance shape, the depth parameter description says do not use it unless the user explicitly asks, the proxy default deliberately avoids installing EnvHttpProxyAgent when no proxy variables exist so a stale variable cannot route traffic through an intermediary returning 403, and stdio mode warns at startup when –image-dir is unset. Credentials come from FIGMA_API_KEY or FIGMA_OAUTH_TOKEN; telemetry can be disabled with –no-telemetry or DO_NOT_TRACK=1 and secrets are redacted from reports. 15.9k stars. It generates no frontend code - the output is design facts, and translating them into React, Vue or SwiftUI is your coding agent job. Not yet run against a Figma file of our own, and no screenshot-versus-structured-data accuracy comparison, so this is graded as pending reproduction.

MCPFigmaDesign-to-Code

TypeScript16k1.3k65

XiaomiMiMo/MiMo-CodeMIT

MiMo Code: Xiaomi terminal coding agent betting on memory and self-evolution

Xiaomi open-sourced terminal-native AI coding assistant, TypeScript built with bun. The source is MIT but usage is additionally bound by USE_RESTRICTIONS.md, the MiMo terms of service and the trademark policy - read those before treating it as plain MIT. The README states it is a fork of OpenCode: it keeps the multi-provider, TUI, LSP, MCP and plugin core and adds persistent memory (SQLite FTS5 full-text search across four kinds - project MEMORY.md, session checkpoints, scratch notes and task progress - injected automatically on session resume), intelligent context management (near the limit it rebuilds from the latest checkpoint plus project memory plus task progress plus retained recent messages, ranked by importance against a token budget), goals and stop conditions (/goal sets the condition, and when the agent wants to stop a separate judge model assesses whether it was truly met, which targets optimistic early quitting), and deterministic JS workflows in a sandbox (compose splits independent tasks into isolated git worktrees with per-task TDD; also deep-research, fact-check with three-reviewer adversarial voting, and research-experiment with an anti-metric-gaming audit). Twenty built-in skills (arxiv, claude-code, codex, docx, pdf, pptx, xlsx, html-to-video, product-design and more), compatible with four skill roots - .agents/skills, .claude/skills, .codex/skills, .opencode/skills - where a user skill of the same name overrides the built-in. /dream and /distill are its signature: the first distils recent session trajectories into project memory and prunes stale entries, the second finds your repeated manual routines and packages high-confidence candidates into reusable skills. Model agnostic - the Xiaomi platform, Codex/ChatGPT OAuth, or any OpenAI-compatible endpoint. 13.4k stars. Not installed or run yet; memory-restore accuracy, judge-model effectiveness and the vendor-stated cache hit rates are unverified, so this is graded as pending reproduction.

Coding AgentAgent HarnessPersistent Memory

TypeScript13k1.4k76

genspark-ai/genofficeApache-2.0

GenOffice: Open-Source AI Office Suite

GenOffice is the first full-featured open-source AI office suite (Apache-2.0): real .docx/.xlsx/.pptx/PDF/Markdown editing with byte-preserving Word compatibility, block-level AI editing with document-aware agents, BYOK model support, fully local.

AI OfficeOpen SourceOffice Suite

TypeScript7.9k1.0k36

Neroued/ninferApache-2.0

NInfer: A From-Scratch C++/CUDA Inference Engine for a Single RTX 5090

A from-scratch C++/CUDA inference engine for five explicitly registered Qwen checkpoints on one NVIDIA GeForce RTX 5090. Startup-frozen residency picks MTP or DFlash speculative decoding, Vision, and one of five KV storage formats; a shared Device KV pool plus pinned Host State/KV checkpoints reuse exact prompt prefixes across 240k-token contexts. Measured aggregate decode reaches 1,146.9 tok/s at concurrency 8 and 15,544.3 tok/s on a 7,680-token prefill.

LLM InferenceCUDASpeculative Decoding

C++1.4k26113

dexmal/dexboticMIT

Dexbotic: A One-Stop VLA Toolbox for Pretraining, Fine-Tuning, Inference, and Evaluation

A one-stop VLA toolbox from Dexmal: the unified DexData format spans pretraining, fine-tuning (full/LoRA/RL), inference, and evaluation, supporting mainstream models (π0, CogACT, OFT, MemVLA, GR00T N1) plus the in-house dual-expert DM0; DB-pretraining brings consistent gains across five simulation benchmarks, with 62% average success on RoboChallenge Table30 real-robot evaluation.

VLAManipulationNavigation

Python1.4k1814

aliyun/ai-agent-handbookApache-2.0

AI Agent HandBook: Alibaba Cloud Enterprise Agent White Paper

Alibaba Cloud open-source (Apache-2.0) white paper distilling enterprise agent deployment experience across the agent lifecycle — architecture, building, runtime, governance and optimization — in 30 Markdown chapters. Core thesis: understand the Model/Harness responsibility boundary instead of attributing everything to model capability. Covers agent loops, task state machines, context builders, action planes, MCP/A2A, sandbox runtimes, AI-gateway governance and evaluation flywheels, plus a 2026 agent developer survey and real cases (ABACI kernel-patch testing, Kitta code review, Bilibili content insights).

AI Agent HandBookAgent HarnessAgentic Application

7041148

dreamers-laboratory/image-to-3d-pipelineApache-2.0

image-to-3d-pipeline: A Fair Benchmark and Browser Explorer for Image-to-3D Models

A fixed-protocol harness for comparing image-to-3D models. One set of AI renders of a fictional submersible goes into TRELLIS V1/V2 and TripoSR; each mesh is inspected through an identical Blender 4.3 EEVEE orbit and scored on a 0-2 five-axis rubric, then the winner ships into a Three.js + three-mesh-bvh explorer with real collision. TRELLIS V2 stochastic ranks first at 1.56 MB / 11,866 faces.

TrellisTripoSRHunyuan3D

JavaScript258200

mekion/the-bimo-projectApache-2.0

Bimo: a 45 cm Open-Source Hip-Head Biped Trained in Isaac Lab and Distilled onto an RP2040

Bimo is a 45 cm, roughly 1.6 kg hip-head biped kit: eight STS-3215 bus servos, a BNO08x 9-DoF IMU, four VL53L0X rangefinders, two 180-degree cameras, and a custom RP2040 board closing a 20 Hz control loop. The repository publishes the Python control API (1023 lines), three MCU firmware builds (1349 lines), a ROS2 wrapper (1431 lines) and an Isaac Lab training environment (874 lines: six reward terms, dense domain randomization, and a system-identified STS3215Actuator with 3.113 Hz bandwidth, directional gear backlash and a 5 ms bus delay), plus a 26MB Bimo.usd model. The CPG gait's 104 Fourier coefficients and per-joint amplitude gains are hardcoded in both Python and C, and a [256,128,64] PPO teacher distills into a [64,32] student that goes through onnx2c into the firmware, so the robot walks untethered on the MCU. All code is Apache-2.0, but the CAD and electronics the README promises are still marked coming soon: no STL, STEP, schematic, gerbers or BOM exist in the working tree or in git history, and neither release ships pretrained weights (zero assets) or a DIY assembly manual. Currently v1.1.0 with 198 stars, in pre-order status.

Bipedal robotsHumanoidOpen Source

Python19887

lightorigins/LightNav-0Apache-2.0

LightNav-0: Eliciting VLM Spatial Intelligence for Generalist Embodied Navigation

A compact generalist navigation model from Light Origins: Qwen3-VL-4B backbone + dual-channel pointing + RVQ action tokens. One checkpoint covers instruction following, object navigation and visual tracking, transferring zero-shot across humanoid/quadruped/wheeled/aerial robots. Trained entirely in simulation, Apache-2.0.

LightNav-0LightOriginsNavigation

Python6120

reubenstr/ElectroPup

ElectroPup: a 12-DoF 3D-Printed Open-Source Quadruped

ElectroPup is a 4.5 kg, 12-DoF DIY quadruped: twelve MG4010E-i10v3 CAN BLDC actuators split into four per-leg CAN segments, a Raspberry Pi 5 for compute, and a single ~7,700-line pure-Python service with no ROS in the stack. The repository publishes OnShape CAD, two KiCad v8 boards, a bill of materials, measured weights and power draw (21.6 W suspended, 33.8 W standing), plus MuJoCo simulation, a React Native/Expo browser UI, four Bezier-foot gaits, and a full build guide. The author also documents known gaps: no license file, no contact sensing, and a simulation model without approximate masses or inertias.

Quadruped robotsOpen SourceBLDC Motor

Python3542

As an Amazon Associate, we earn from qualifying purchases.