Disclosure: this article contains no affiliate links. All links are plain links, and we earn nothing if you click them. See our affiliate disclosure.
How this was made: researched and drafted with AI tools from the sources linked in this article, then checked against those sources before publishing.
Quick answer: On November 10, 2026, three AI repos showed up on GitHub’s daily trending list: daily_stock_analysis, an LLM tool that writes a daily report on your stock watchlist; Garry Tan’s GBrain, a self-hosted memory for AI agents; and HA-MCP, which lets Claude or ChatGPT control Home Assistant. All are MIT-licensed; model and API costs are yours.
Star counts below are as of November 10, 2026, about 07:00 MYT, from GitHub Trending (daily). We are not guessing why any repo jumped.
At a glance
On a narrow screen, swipe the table sideways to see every column.
| Repo | What it does | Free to run? | Who it’s for |
|---|---|---|---|
| daily_stock_analysis (68,790★, +505 today) | Builds a daily AI report on a stock watchlist | Code is free (MIT). README says you can run it on GitHub Actions with no server. Default market data is free; you still need an LLM API key or a local model | People who want a daily summary of stocks they already follow |
| gbrain (32,104★, +215 today) | Gives the agent you already use a memory you control | MIT and free to self-host. gbrain.io is optional paid hosting; we did not verify its price | Heavy agent users and small teams |
| ha-mcp (5,911★, +31 today) | 87 MCP tools to configure and automate Home Assistant | Free. Needs Home Assistant 2026.8.0 or newer, plus an assistant that supports MCP | People already running Home Assistant |

daily_stock_analysis: a daily watchlist report
Per the README, this LLM system pulls quotes, news and fundamentals, builds a “decision dashboard,” and can push updates to WeCom, Feishu, Telegram, Discord, Slack or email.

The author recommends GitHub Actions: “5 minutes to deploy, zero cost, no server.” Default data sources include AkShare, Baostock and YFinance, which can run without extra tokens. Those free sources, the README says, can be rate-limited or unstable. You still need at least one LLM API key (Gemini, Claude, an OpenAI-compatible API, or a local Ollama model, among others). That bill is yours.
The README disclaimer is clear: the project is “for learning and research only” and “does not constitute any investment advice.” We are not recommending third-party API resellers listed as sponsors on the page.
Latest push we saw: November 9, 2026, 22:14 MYT.
GBrain: portable memory for your agents
Garry Tan’s GBrain stores facts with sources and corrections, then shares them across agents over MCP. You can start with keyword retrieval that needs no extra keys, then layer on optional semantic search and synthesis.

It is MIT-licensed and free to self-host. The README mentions gbrain.io as optional paid hosting; we did not check that price, so we are not quoting one.
The “155,795 pages, 24,589 people, 5,340 companies” figures on the README are Garry Tan’s own stats for his setup. The docs mention a --funnel flag for cloud agents such as Grok Bot, Muse and ChatGPT. Read the project’s security notes before you expose a memory store over a network.
Latest push we saw: November 10, 2026, 06:56 MYT.
HA-MCP: talk to Home Assistant through Claude or ChatGPT
HA-MCP is an MCP server with 87 tools for configuring and automating Home Assistant from Claude, ChatGPT and other assistants. Recommended install: the HA-MCP integration through HACS or, on Home Assistant 2026.11 or newer, the Marketplace. It needs Home Assistant 2026.8.0 or newer.

Setup asks for an administrator access token. An administrator token is recommended; non-admin tokens work with limitations. A v7.3.0 breaking change moved ha_config_set_yaml to beta. Read the README before you give any assistant control of your home.
This pick had a weaker daily gain than the other two. We left it in because it is a useful AI tool that we have not featured recently, and no stronger unused AI repo beat it on this morning’s list.
Latest push we saw: November 10, 2026, 06:34 MYT.
Which one fits you
Want a scheduled market briefing and you already have an LLM key? Start with daily_stock_analysis, and treat every output as research, not advice. Running several coding agents and tired of repeating context? Look at GBrain. Already on Home Assistant and want to talk to it from Claude or ChatGPT? HA-MCP is the direct fit.
Still trending, covered earlier
Astryx and Claude Quickstarts were still high on today’s list; we covered them in yesterday’s trending AI repos. OpenMausBot was in the November 8 daily post. Agent Browser was in the November 6 daily post. For a wider view, see this week’s top 10.
Related on TechKit: GitHub Trending AI repos, Nov 9 · This week's top 10
Quick questions
Are they free?
The code is MIT-licensed. You pay for your own LLM API or assistant subscription. GBrain’s hosted option, gbrain.io, is described as paid; we did not verify the price.
Can daily_stock_analysis run without a server?
The README says yes, via GitHub Actions.
Is the stock analysis investment advice?
No. The README says it is for learning and research only and does not constitute investment advice.
Is HA-MCP safe?
It asks for an administrator access token, and an admin token is recommended. Follow the README’s security notes before exposing it.
Why does the list change daily?
GitHub ranks repositories by recent star growth, so the order can move a lot from one day to the next.
Sources
– GitHub Trending, daily view (checked November 10, 2026, about 07:00 MYT)
– ZhuLinsen/daily_stock_analysis
Last checked: November 10, 2026, 07:00 MYT.
1 thought on “Today’s Trending AI Repos on GitHub (Nov 10, 2026): Stock Analysis, GBrain & HA-MCP”