AI and LLM Tools
The largest cluster of trending repos today centers on AI agents, LLM access, and voice generation.
- diegosouzapw/OmniRoute -- A free MIT-licensed AI gateway routing requests across 268+ providers and 500+ models with quota-aware fallback and token compression. (+1343)
- jamiepine/voicebox -- Open-source AI voice studio for cloning, dictating, and creating voice content. (+610)
- kvcache-ai/ktransformers -- A flexible framework for heterogeneous LLM inference and fine-tune optimizations. (+360)
- bojieli/ai-agent-book -- Open-source book on AI agent design principles and engineering practice, with full text and code. (+1734)
- rohitg00/ai-engineering-from-scratch -- A structured resource for learning, building, and shipping AI engineering projects. (+501)
- ruvnet/ruflo -- A meta-harness for deploying multi-agent swarms with adaptive memory, RAG integration, and support for multiple coding agents. (+154)
- elder-plinius/G0DM0D3 -- A liberated AI chat interface. (+216)
- MoonshotAI/kimi-cli -- A CLI agent powered by Kimi, designed as a next-generation command-line coding assistant. (+410)
- topoteretes/cognee -- Open-source AI memory platform giving agents persistent long-term memory via a self-hosted knowledge graph. (+303)
- KnockOutEZ/wigolo -- Local-first search, fetch, crawl, and research over MCP for AI coding agents, with no API keys and no cloud cost. (+595)
Agentic Coding and Dev Workflow
A strong theme today is tooling that wraps, orchestrates, or supercharges coding agents.
- garrytan/gstack -- A curated Claude Code setup with 23 opinionated tools covering engineering, design, QA, and release management roles. (+330)
- traycerai/traycer -- A nerve center for agentic coding workflows. (+20)
- 1jehuang/jcode -- A coding agent harness written in Rust. (+235)
- openinterpreter/openinterpreter -- A coding agent targeting open models like Kimi K3, rewritten in Rust. (+176)
- openclaw/openclaw -- A personal AI assistant that runs on any OS and any platform. (+161)
- Gitlawb/openclaude -- A flexible Claude runner described as runs anywhere, uses anything. (+31)
- tirth8205/code-review-graph -- Local-first code intelligence graph for MCP and CLI that reduces context sent to AI tools during reviews. (+663)
- cocoindex-io/cocoindex -- An incremental processing engine designed for long-horizon agents. (+54)
Observability and Infrastructure
- SigNoz/signoz -- OpenTelemetry-native observability platform covering logs, metrics, traces, APM, and distributed tracing, now with MCP and AI agent support. (+345)
- PostHog/posthog -- Full product analytics and observability platform with AI observability, session replay, feature flags, and error tracking. (+411)
- tursodatabase/turso -- A SQLite-compatible SQL database in Rust that is adding experimental Postgres support. (+60)
- Q00/ouroboros -- An Agent OS focused on specification over prompting. (+17)
Web Frameworks and UI
- tldraw/tldraw -- React SDK for building infinite canvas applications, positioned as agent-friendly. (+73)
- ibelick/ui-skills -- A skills and reference resource for design engineers. (+536)
- firecrawl/open-lovable -- Clones and recreates any website as a modern React app using Firecrawl. (+34)
- every-app/open-seo -- An open-source alternative to Semrush and Ahrefs for SEO analysis. (+222)
Productivity and General Tools
- super-productivity/super-productivity -- Advanced todo and time-tracking app with Jira, GitLab, and GitHub integrations. (+30)
- Canner/WrenAI -- Generative BI platform for text-to-SQL across 20+ data sources including BigQuery, Snowflake, and Redshift. (+121)
- HKUDS/DeepTutor -- Lifelong personalized tutoring system. (+269)
- Comfy-Org/ComfyUI -- Modular diffusion model GUI and API with a graph-nodes interface. (+138)
- Robbyant/lingbot-map -- A feed-forward 3D foundation model for reconstructing scenes from streaming data. (+865)
- pollinations/pollinations -- An open-source generative AI platform. (+6)
Today's trending list makes one pattern hard to ignore: builders are layering agent orchestration, local-first tooling, and LLM gateway infrastructure at a pace that suggests the scaffolding around AI coding agents is becoming its own category. The simultaneous rise of memory platforms, context compression tools, and multi-provider gateways points to a community actively solving the operational costs of running agents at scale.