The agentic repositories, by stars and by pulse
45,094 repositories that exist to build agents — servers, agent apps, skill packs, RAG stacks — 12,381 of them pushed to in the last week. The largest, the fastest-gaining, the most recently touched and the newest, with forks, open issues, language and role on every row, read from GitHub every night.
45,094 repositories · read 2026-09-17 · 7 days of star history since 2026-09-10Repositories created per day
By GitHub's creation date, for repositories that later made it into this corpus. The right edge trails discovery by a few days.
| # | Repository | Stars gained | Change | Stars | Forks |
|---|---|---|---|---|---|
| 1 | ayghri/i-have-adhd A skill to stop your coding agent from burying the answer. ADHD-friendly output. | +49.2%4,809/day | 44k | 2,747 | |
| 2 | deepseek-ai/deepseek-harness DeepSeek Harness: Everything is a Plugin. | +2.7%1,957/day | 222k | 27k | |
| 3 | tt-a1i/archify Agent skill for beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—self-contained HTML with motion and crisp export. | +10.4%1,886/day | 60k | 4,350 | |
| 4 | DietrichGebert/ponytail Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. | +3.7%1,644/day | 137k | 7,558 | |
| 5 | cathrynlavery/diagram-design 38 editorial diagram types for Claude Code, Codex, and Pi. Self-contained HTML + SVG. No shadows. No Mermaid slop. | +13.6%1,557/day | 39k | 2,592 | |
| 6 | affaan-m/ECC The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond. | +1.3%1,114/day | 257k | 39k | |
| 7 | alibaba/open-code-review Fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in multi-language ruleset (NPE, thread-safety, XSS, SQL injection), OpenAI & Anthropic compatible. | +11.0%3,318/day | 33k | 2,375 | |
| 8 | stablyai/orca Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and remote runtime. | +5.1%1,086/day | 67k | 4,625 | |
| 9 | alibaba/open-code-review Fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in multi-language ruleset (NPE, thread-safety, XSS, SQL injection), OpenAI & Anthropic compatible. | +11.7%3,160/day | 30k | 2,375 | |
| 10 | alibaba/open-code-review Fast, efficient, battle-tested at Alibaba's scale. Hybrid architecture code review tool: deterministic pipelines + LLM Agent, precise line-level comments, built-in multi-language ruleset (NPE, thread-safety, XSS, SQL injection), OpenAI & Anthropic compatible. | +11.4%2,745/day | 27k | 2,375 | |
| 11 | diegosouzapw/OmniRoute Never stop coding. Free MIT AI gateway: one endpoint, 352 providers (150+ free), 1200+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 550+ contributors | +4.1%861/day | 65k | 9,459 | |
| 12 | blader/humanizer Agent skill that removes signs of AI-generated writing from text | +4.7%714/day | 47k | 4,002 | |
| 13 | vastsa/PI-Desktop Local-first AI coding agent desktop: Electron + Rust host core + pi Agent Harness + user-installable plugins | +170%669/day | 3,185 | 341 | |
| 14 | DietrichGebert/ponytail Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. | +1.4%976/day | 141k | 7,558 | |
| 15 | alsk1992/CloddsBot Open Source AI trading agent that operates autonomously across 1000+ markets - Polymarket, Kalshi, Binance, Hyperliquid, Solana DEXs, 5 EVM chains. Scans for edge, executes instantly, manages risk while you sleep. Agent commerce protocol for machine-to-machine payments. Self-hosted. Built on Claude. | +193%578/day | 2,634 | 335 | |
| 16 | firecrawl/firecrawl The context API to search, scrape, and interact with the web at scale. 🔥 | +1.0%578/day | 180k | 9,830 | |
| 17 | heygen-com/hyperframes Write HTML. Render video. Built for agents. | +3.5%560/day | 49k | 4,639 | |
| 18 | affaan-m/ECC The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond. | +0.6%825/day | 260k | 39k | |
| 19 | NousResearch/hermes-agent The agent that grows with you | +0.7%537/day | 245k | 52k | |
| 20 | arcboxlabs/arcbox Run AI agents on real and isolated machines — own kernel, filesystem, and network — with <100ms boot. Local first, OCI compatible, pure Rust. | +37.3%509/day | 5,619 | 173 |
Two readings of the same repository, subtracted, over the days between them — the rate is per day so a two-day window and a ten-day window compare. Repositories under 50 stars are excluded so the board is about momentum rather than a first week.
How the corpus is made
Role is inferred from the repository's own files — a server manifest, a skill directory, an agent framework import. Pulse is the last push date as read last night. Archived repositories and forks are excluded from every list on this page.
Which repositories ship the most skills, and which skill names are becoming a convention, is in The agentic layer, by structure. The skills themselves have their own board: Skills.