The GitHub repos that hit the daily top 10
Every day we count the stars on every repository we list: MCP servers, agent skills, the AI-stack products on our layer pages and the open-source apps directory. The ten that gained the most since the day before make that day’s board, and the board beside this ranks today, this week, this month or all time by stars gained over that window.
Below is everyone who has made a daily top ten, with the day-by-day record beside each one. Filter by what it is, what it is for, and how high it got. One strong day and a steady run look different on the strip: that difference is why the strip is there.
Everyone who has been in the top 10 ↓ · How it is worked out
Stars gained 20 Oct → 27 Oct
- #1VoiceStudiodebpalash · AI stack · GitHub ↗+11,901
- #2ponytailDietrichGebert · AI stack · GitHub ↗+9,102
- #3orcastablyai · AI stack · GitHub ↗+8,719
- #4graphifyGraphify-Labs · AI stack · GitHub ↗+7,027
- #5skillsmattpocock · Agent skills · GitHub ↗+6,526
- #6xerjxerj-org · AI stack · GitHub ↗+6,233
- #7ECCaffaan-m · AI stack · GitHub ↗+5,770
- #8iFixAiifixai-ai · AI stack · GitHub ↗+5,007
- #9freellmapitashfeenahmed · AI stack · GitHub ↗+4,525
- #10ghidraNationalSecurityAgency · Open-source app · GitHub ↗+3,093
Everyone who has been in the top 10
Every repository that placed in a daily top ten this week. Sorted by stars gained over the window; change the sort for other orders.
- Placed
- 18
- Been #1
- 3
- Kinds
- 4
- Days ranked
- 7
18 of 18 repositories
- best#1
debpalash/VoiceStudio
VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.
- Gained · This week
- +11,901
- Stars
- 89,738
- Days at #1
- 4
- Days in top 10
- 7
- Biggest day
- +2,493
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated5351111 - best#1
DietrichGebert/ponytail
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
- Gained · This week
- +9,102
- Stars
- 177,372
- Days at #1
- 2
- Days in top 10
- 7
- Biggest day
- +1,900
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated2112785 - best#2
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.
- Gained · This week
- +8,719
- Stars
- 107,291
- Days at #1
- —
- Days in top 10
- 7
- Biggest day
- +1,599
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated4224324 - best#3
Graphify-Labs/graphify
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, e
- Gained · This week
- +7,027
- Stars
- 136,324
- Days at #1
- —
- Days in top 10
- 7
- Biggest day
- +1,373
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated64461037 - best#5
A pack of 39 agent skills.
- Gained · This week
- +6,526
- Stars
- 304,239
- Days at #1
- —
- Days in top 10
- 7
- Biggest day
- +1,183
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated7665878 - best#2
xerj-org/xerj
XERJ is the new way for AI to search data. Its autoindex capability activates agents to know your data without the token waste of grep and sed. One command indexes code, docs, logs and PDFs for search
- Gained · This week
- +6,233
- Stars
- 16,904
- Days at #1
- —
- Days in top 10
- 4
- Biggest day
- +1,828
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated3323 - best#5
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.
- Gained · This week
- +5,770
- Stars
- 290,031
- Days at #1
- —
- Days in top 10
- 7
- Biggest day
- +1,043
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated8777566 - best#3
ifixai-ai/iFixAi
Independent Auditing of AI Agents. Run by human or the agent itself, to answer the most crucial question in the AI Agent Economy. Is the agent doing what is supposed to do? With iFixAi you can have th
- Gained · This week
- +5,007
- Stars
- 35,201
- Days at #1
- —
- Days in top 10
- 5
- Biggest day
- +1,320
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated5391010 - best#6
tashfeenahmed/freellmapi
7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart routing, automatic failover, encrypted key
- Gained · This week
- +4,525
- Stars
- 41,675
- Days at #1
- —
- Days in top 10
- 6
- Biggest day
- +730
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated988869 - best#10
NationalSecurityAgency/ghidra
Ghidra is a software reverse engineering (SRE) framework
- Gained · This week
- +3,093
- Stars
- 93,262
- Days at #1
- —
- Days in top 10
- 1
- Biggest day
- +537
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated10 - best#2
HKUDS/DeepTutor
DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.
- Gained · This week
- +2,419
- Stars
- 44,243
- Days at #1
- —
- Days in top 10
- 1
- Biggest day
- +1,422
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated2 - best#10
akitaonrails/ai-memory
Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors
- Gained · This week
- +2,358
- Stars
- 12,476
- Days at #1
- —
- Days in top 10
- 1
- Biggest day
- +517
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated10 - best#9
supabase/supabase
The Postgres development platform. Supabase gives you a dedicated Postgres database to build your web, mobile, and AI applications.
- Gained · This week
- +2,150
- Stars
- 113,137
- Days at #1
- —
- Days in top 10
- 2
- Biggest day
- +616
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated109 - best#4
coollabsio/coolify
An open-source, self-hostable PaaS alternative to Vercel, Heroku & Netlify that lets you easily deploy static sites, databases, full-stack applications and 280+ one-click services on your own servers.
- Gained · This week
- +1,727
- Stars
- 64,599
- Days at #1
- —
- Days in top 10
- 1
- Biggest day
- +841
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated4 - best#9
gitroomhq/postiz-app
📨 The ultimate agentic social media scheduling tool 🤖
- Gained · This week
- +1,204
- Stars
- 38,631
- Days at #1
- —
- Days in top 10
- 1
- Biggest day
- +813
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated9 - best#9
google-gemini/gemini-cli
An open-source AI agent that brings the power of Gemini directly into your terminal.
- Gained · This week
- +1,100
- Stars
- 108,192
- Days at #1
- —
- Days in top 10
- 1
- Biggest day
- +620
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated9 - best#1
MadsLorentzen/ai-job-search
The job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
- Gained · This week
- —
- Stars
- 61,194
- Days at #1
- 1
- Days in top 10
- 1
- Biggest day
- +3,241
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated1 - best#4ai-job-search-dev/ai-job-searchRepository
○ no description published
- Gained · This week
- —
- Stars
- 65,812
- Days at #1
- —
- Days in top 10
- 4
- Biggest day
- +836
Daily rank · 21 Oct to 27 Octstronger colour = higher rank · stripes = estimated91045
How the ranking is worked out
- One reading a day. Star counts are read fresh from GitHub once a day for every repository we publish: the MCP server, agent skills and open-source apps directories, and every live product on the layer pages that names its repository.
- A day’s gain is today minus yesterday. If the day before is missing, that day is not ranked for that repository. Two days of stars are never counted as one.
- Earlier days are estimated, and striped. Our own daily readings began on 25 September 2026. For the days before that, back to 1 September, and for days a reading was missed, the day-by-day shape comes from OSS Insight’s daily star counts. Between two of our readings their shape is scaled to match what we measured at both ends; before our first reading it is used as it is. Their figures run roughly 20% either way on a busy repository and more on a quiet one, so a striped rank is a good guide and not a measurement. Before 1 September 2026 there is no usable daily record, so the 90-day view is blank there rather than guessed.
- A week, a month or all time is one subtraction. The count on the latest full reading minus the count at the start of the window, so a missed day does not lose the stars gained on it. A “full” reading covers at least half the repositories the best-covered day does. When the first full reading is later than the window’s start, the board prints the shorter span it really measured; it is never scaled up to the full length. Our first full reading is 25 September 2026, so for now “this month” and “all time” both start there.
- A custom range starts and ends on full readings only. The two day pickers offer just the days we read for most repositories, so a range cannot begin on a day that holds only the estimated big ones. The ranking is the same subtraction, between the two days you pick.
- Only gains place. A repository that stayed flat or lost stars did not place, however many stars it has.
- This is not GitHub’s trending page. That page ranks all of GitHub with a formula GitHub does not publish. This one covers only the repositories we list and uses one rule, stated here.
- Stars are not quality. A launch post can win a day. The strip shows whether a repository held its place or had one big day.
Related
For what each server can reach and the credential it wants, see the MCP server directory. For instructions a model loads, see agent skills.