tigusigalpa / ai-kit
AI-KIT is a practical starter kit for development: ChatGPT/Codex, Claude, DeepSeek, KIMI, and other coding agents. It provides reusable project context, rules for each stack, adaptive model routing, conservative installation and updates, testing and review workflows, security guidance, Docker and Kubernetes readiness, and clean project templates.
RollingGo-AI / RollingGo-hotel-MCP-CN
RollingGo官方酒店MCP，200万酒店实时库存直连，11万直签，渠道专属价，兼容40+主流大模型代理。企业和个人开发者均可免费申请，无调用上限。可按地点、日期、星级、人数、标签等条件搜索酒店，返回候选酒店及展示价。可查询指定酒店的实时房型列表、价格计划和取消政策。
MichaelS1011 / ephemora-cell
Ephemeral, stateless, capability-bound execution for untrusted AI and agent-generated code.
Skyltliu / LocoAgent
A lightweight coding agent 🧑💻 built from scratch, with tool calling, session persistence, workspace context, memory, and multi-provider LLM support 🛠️.
PyModel / jev-judge-mcp
Typed judgment tools for MCP agents. TypeSafe's Jev model as verify, screen, find, classify, rerank, decide, compare, extract, review, gate, and score: the model judges, policy decides auto, review, or escalate.
Ltamann / ComfyUI-H3-Motion-Context-Auto-Chain-addon
ComfyUI-H3-Motion-Context-Auto-Chain-addon
SatoDive / Minimax-H3-Latent-Continuation
This suite enables **long, seamless multi-shot video continuations** with **100% visual context and environment consistency**—completely eliminating generation-to-generation quality degradation!
kyungseo / learnstead
Local LLM·RAG·AI Agent·바이브 코딩을 개념부터 실습과 검증까지 다루는 학습 자료
lixiaochuan2020 / agentic-context-management
This is the official repository for the paper "ACM: Agentic Context Management for Long-Horizon Tasks" [https://arxiv.org/abs/2607.23809]
sauravsingla / agentweave
Pre-inference tool and agent routing for MCP, A2A and tool-rich LLM systems. Reduce candidate tools before model inference with policy-aware routing and recovery.
00200200 / usagetrim
Fast local CLI and MCP tool for compacting verbose test runners and build output.
Qin-sor / agent-context-lab
A minimal coding-agent harness built to measure what context management actually costs: constraint survival after compaction (8%→83%), default-FAIL verification against false completions, and KV-cache prefix design (21%→90%). Stdlib-only, real deepseek-chat runs
Astral-xm / ComfyUI-MiniMax-ContextIR
ComfyUI MiniMax H3 全家桶：Unified to Video / Context IR / Multimodal Chat / Audio Lock / Media Loader / Resolution Selector / Concat AV 等 8 个节点，覆盖提示词优化到本地采样完整工作流
EverlastingAI-Official / LifeContextOS
It is a "LifeContext" operating system designed to run entirely on-device, transforming personal data—such as chat logs, documents, and oral histories—into digital memories and personality structures that are traceable, editable, and portable. We aim to establish a standardized LifeContext format, creating an open foundation for personalized agents
smartcmd / codex-context-window
Codex skill and CLI utility for safely configuring model context windows, effective percentages, and auto-compaction defaults.
adpanru / cordis-mini
用约 600 行纯 Python，从零实现 Cordis / deepseek-harness 的核心 Agent 插件架构：插件系统、Context 服务注册、依赖拓扑排序、事件分发、Waterfall 中间件与可逆副作用。无需真实 LLM API，内置 Fake LLM，通过完整 Agent Loop 直观理解 AI Agent 插件架构。
seanyourhighness / vllm-sm12x-nvfp4-dflash2
All-NVFP4 vLLM + DFlash2 K7 for Blackwell (SM120/SM121): NVFP4 target/draft/KV, 262K context, optional CPU vision sidecar.
trinity1994 / MCP-Server-Aster-DEX
MCP server for Aster DEX — a Model Context Protocol server that gives Claude, Cursor and Cline structured access to Aster DEX market data, volume, points and airdrop eligibility via natural language. Read-only tools — no trading, no keys. Unofficial community project, not affiliated with Aster.
thaddeuswycliffe / glm-5-3-api-guide
A focused Argolink guide for GLM 5.3: the exact model ID, chat-compatible request, reasoning controls, long-context limits, and production error handling.
agent-clinic / claude-md-doctor
Give your CLAUDE.md / AGENTS.md a checkup — audit size vitals, dead references, drifted claims, and backtest every rule against your own session history to see which rules get followed, ignored, or never used. A doctor-style report that cites its evidence.
RankSpotAI / awesome-seo-mcp
A curated list of MCP servers for SEO. Search Console, keywords, backlinks, crawling, SERPs and AI search visibility, verified with install details and auth requirements.
isabelpapaya / MCP-Server-Lighter
MCP server for Lighter — a Model Context Protocol server that lets Claude, Cursor and Cline query Lighter (lighter.xyz) volume, points, rank and airdrop eligibility in natural language. Read-only — public endpoints only, no keys. Unofficial community project, not affiliated with Lighter.
lucidrains / bdh-cq
Implementation of BDH-CQ: In-Context Learning with Recurrent Latent Reasoning, proposed by Pathway Research
m2dumpling / pawbot
A self-hosted AI agent runtime with WebUI, MCP tools, context governance, and Record & Replay for deterministic execution & evaluation.
Emmimal / intent-continuity
A pure-Python system that automatically discovers, verifies, and applies relevant historical requirements for coding agents — no embeddings, no vector database, no LLM calls. Companion code for the Towards Data Science article on intent continuity.
icyberdeveloper / tbank-mcp
Unofficial MCP server for T-Bank (Т-Банк, ex-Tinkoff) mobile banking: accounts, cards, transfers, bill pay, grocery ordering, tickets & travel — for Claude Code and any MCP client.
benogoulart / GitWhisper
Auto-generate Conventional Commits from your staged diff — detects scope, type, and context. Cross-platform (Windows/Linux/macOS). No dependencies.
nibor1896 / Crow
Crow — a local AI agent, not a chat box: 30 tools, MCP, memory, skills, goals, subagents, vision, and image generation and editing with Qwen-Image. Runs on crow-nest or llama.cpp on one GPU, or any model via OpenRouter. Windows and Linux.
RinDig / cost-of-remembering
The Cost of Remembering: filesystem memory matches long-context accuracy on LongMemEval while reading 97% fewer tokens and costing 95% less. Harness, run data, 129 agent-built memories, and paper source.
ismailsaoulaj / reddit-mcp-server
A production-grade Model Context Protocol (MCP) server. Empowers AI assistants (Claude, Cursor) to search, fetch, and analyze Reddit discussions with resilient rate-limiting and smart LLM noise filtering.
polaris-smart / mcd-agent-gourmet
把菜单价换成官方当次核价：一句话说清预算和口味，给你一个答案 + 可复制的点单卡，并算清比另一种点法省多少钱（示例：单品直选 ¥48.50 vs 三件套 ¥36.50，省 ¥12.00/24.7%）。只读 MCP、零依赖 Python、免 Token 网页体验。
supElement / ComfyUI_MinimaxH3_AutoContext
A long video generation solution that breaks through the memory barrier: automatic segmented inference + inter-segment Latent anchoring, allowing MiniMax H3 to generate infinitely continuous long videos with limited video memory.
ruashots / ComfyUI-OpenH3-IR
Four ComfyUI nodes for MiniMax H3: name your media with @, lock exact dialogue, and get the structured brief H3 wants. No <Picture 1> bookkeeping, no second service to start.
zhihz / openjev
Local bilingual probability decisions from context, questions, and candidate answers. Independent research preview inspired by TypeSafe Jev.
mikeshu2001 / personal-brain-builder
Local-first agent workflow for building private cross-project memory with explicit consent, validation, and crash-safe history.
flamexnreal / davinci-resolve-ai-bridge-mcp
DaVinci Resolve (Free Version and Studio) MCP server for Claude, Cursor, Codex, and Antigravity. Full timeline editing, cuts, camera zoom, and color grading.
tonyd2wild / DS4-H3-Video-Gen-Factory
Run DeepSeek-V4-Flash at full 1M context AND two MiniMax H3 video instances on the same two DGX Sparks. Benchmarked C1-C6.
webkubor / claude-usage-statusline
Claude Code 状态栏：会话花费、上下文占用、5h / 7d 限额预警，区分订阅制与按量付费。单文件 Python，零依赖。· A Claude Code statusline for session cost, context usage and limit warnings.
Intelligible / elbi
Elbi gives your data agent a verified fact store.
sepehr071 / snapp-mcp
Unofficial read-only MCP server for Snappfood and SnappMarket (اسنپفود), Iran: search food and groceries near you, compare prices, find deals. For Claude, Cursor and Copilot via uvx.
sepehr071 / digikala-mcp
Unofficial read-only MCP server for Digikala (دیجیکالا), Iran: search products, compare every seller and price, price history, reviews and deals. For Claude, Cursor and Copilot via uvx.
NirDiamant / Agentic_Engineering
Your coding agent starts every session knowing nothing about your project. One line, and it writes its own memory from your real code, then grades itself.
shriramkv / mcp-load-lab
Load and latency harness for Model Context Protocol (MCP) servers. Drives real tools/call traffic at configurable concurrency, measures latency percentiles and throughput, and fails CI when a server misses its performance budget. Stdlib-first, zero required dependencies for stdio.
DocPang / qwen38-k100ai-int8-optimization
Qwen3.8-27B INT8/W8A8 optimization for Hygon K100AI: SGLang, TP1/TP4, DFlash2 and long-context Agent workloads
superhero-tech / superhero-skills
Battle-tested agent skills for Product Builders. Part of the Product Builder OS by Superhero.tech: Super Brain for context, skills for action, built from 32+ years of combined product experience and pressure-tested with 1,800 builders.
fabricioveronez / prompt-e-contexto-na-pratica
Exemplos didáticos de engenharia de prompt e engenharia de contexto em Python (LangChain + Streamlit) — veronez.io
dg1kjd / sglang-sxm2
Agentic-coding oriented inference engine for N x V100-SXM2, based on SGLang: GLM-5.3-Flash, Qwen3.8-Flash-Next, DeepSeek-V4.1-Flash, MiniMax-H3.
BeamManP / mmd-mcp
Windows stdio MCP server for MikuMikuDance authoring and MME settings
luoyu100 / PoS
An inference-time framework for long-horizon LLM agents with explicit belief states, consistency validation, and trapping-aware recovery.
satellitedown / fast-long-context
Easy local setup for Qwen3.8-27B Abliterated: 256K context and measured 240-300 tokens/s generation on an RTX 5090.
dotdna-io / assetatlas
Open-source asset inventory — turn scattered, contradictory tool data into one reliable inventory, then chat with it using the LLM of your choice. Tagging, posture rules, and alerting built in.
ZhuWenjie98 / MMOOC
MMOOC: A Comprehensive Benchmark for Out-of-Context Evaluation in Multimodal Large Language Models
eater2 / ai_agents_api_library
Verified catalog of 300+ APIs and MCP servers with opinions and statistics so that AI agents can call after a one-time API key/OAuth setup. JSON, llms.txt and an MCP server for agents.
HorizunGroup / horizun-pbi-mcp
Open-source Power BI MCP to build, audit and repair Desktop models and PBIP reports with DAX, TOM, TMDL and PBIR.
jnardiello / GLM-5.3-Flash-FP8-4-DGX-Spark-Switchless
GLM-5.3-Flash (FP8) + DFlash2 speculative decoding on 4x NVIDIA DGX Spark / ASUS GX10: vLLM TP4, 256K context, switchless ConnectX-7 ring with patched NCCL
nisesen / inferenceinquire
Real-time monitoring for llama.cpp inference servers: a read-only host agent plus a web dashboard for hardware, health and inference telemetry (prefill/decode vs context depth, slot occupancy, KV cache, GPU and PCIe).
SmetDenis / openwebui-token-usage-display
A filter plugin for Open WebUI that shows detailed token, context, timing and cost statistics below each AI response: token counts (input/output/total, reasoning, cached, audio), context-window utilization, generation time, tokens/second and message/chat cost.
drowzeys / keys-vLLm.0.27.1-GLM-5.3-Flash-NVFP4-NVFP4KV-1M-Context-Abliterated
GLM-5.3-Flash (Z.ai) NVFP4 (LibertAI) + NVFP4 KV, 1M context on 2x DGX Spark GB10 — vLLM TP2 recipe, walls ledger, benchmarks, needle to 1.04M; abliteration with blackforest_AI next
luxingcom / LuZ-0.1.7-DeepSeek-v4.1-Flash-DGXspark-TP4-Ring
DeepSeek-V4.1-Flash on 4x DGX Spark TP4 switchless RoCE ring - SGLang production recipe, 600K context, DSpark MTP, benchmark archive
Adolanium / hermes-plugin-shodan
Shodan.io for Hermes Agent. Host intel, search, DNS, CVE lookups, monitoring and on-demand scanning, with credit budgeting and response shaping that keeps results inside the context window.