Opyt

Your attention is the context your AI is missing.

Every bookmark, follow and subscription was you deciding whose thinking is worth your time. Opyt turns that into a knowledge base that keeps growing. It pulls in new work from the people you already read and the topics you keep coming back to.

Sources

It runs on what you already do.

Point Opyt at the sources you already use: X, Substack, GitHub, research papers, blogs, any URL you hand it. The saving, following and subscribing you already do is the input, and everything they publish ends up in one place you can search.

  1. 01 curate onboard

    Everything you saved before today is searchable the first time you run it. You are not starting from an empty store.

    you › Pull in what I've been saving.

    onboard()
    x bookmarks 691 saved posts, full text x likes 99 liked authors x follows 492 accounts x lists 6 lists substack 19 subscriptions · 12 saved posts, full text read from your logged-in browser · no API key, no password

    claude › Everything you saved is in. 703 posts are now searchable, and Opyt knows the 1,193 accounts they came from.

  2. 02 screen oracle

    You do not have to remember who is worth listening to. Your platform engagement already records it, and Opyt counts it up.

    1@jasonfurmanyou follow · you subscribe · saved 12×
    2@drvoltsyou follow · saved 8×
    3@karpathyliked 4×
  3. 03 track add_oracle

    Opyt pulls someone's whole archive, going back years, not the three posts you happened to bookmark.

    You confirmed @karpathy on X, and Opyt found and verified his other platforms on its own.
    3,412posts from X, in full text
    38repos found on GitHub, READMEs included
    96essays found on his own site

What it reads

Wherever you read, one store.

Xbookmarks · likes · follows · Listsyour browser session
Substacksubscriptions · saved posts, full textyour browser session
GitHuba tracked person's repos and READMEspublic
Blogsa tracked person's whole archive, found by feed or sitemappublic
arXiv · OpenAlexpapers matching your standing questionspublic
Any URLa paper, post, article or repo you hand it, saved in fullyou
LinkedInposts and articles from the people you trackcoming soon
AggregatorsHacker News · Hugging Face · npm · crates.io · Stack Overflow · Product Hunt · Homebrewcoming soon
Fundingcompany and round data, asked by namecoming soon

Automatic growth

Your topics keep growing without you.

Interact with your topics the way you already do with AI. Opyt organizes what you save, reads a topic through, and writes its own questions from it, then keeps them running against arXiv, GitHub and OpenAlex. New work on your topics is already there when you go looking, with nothing for you to set up or refresh.

04 grow sitting

Your library gets better on the days you never open it.

March to May

Eleven things about how agents should keep memory arrive in your store: bookmarks, a Substack thread, two papers. You were reading, not filing.

May

you › Read everything I've saved on how agents keep memory.

sitting(action="read")

claude › Ninety-six items, fourteen authors, oldest first. That read left three questions running, none typed by you. Here is one:

Is an agent that edits its own memory more reliable over a long run than one handed a fixed context?
August · you ran nothing in between

claude › That question caught letta-ai/letta, agents whose memory edits itself. It is in your store now, and the topic is seven items bigger than the day you read it.

05discoverfrontier
Tuesday

you › Read everything I've saved on data-center grid costs.

codex › Read the whole topic end to end, oldest first. Three questions came out of that read and they are still running. You typed none of them:

Who ends up paying for data-center grid upgrades? What moves retail electricity prices besides fuel? Do infrastructure costs pass through to household bills?
Friday · you ran nothing in between

you › Anything new?

codex › Six things came in since Tuesday, ranked by how many of your own questions reached each one:

Tue 1paper · Who Pays for the Load? Data-Center Demand and Retail Ratesall 3 questions · 2 of your topics Wed 2repo · PyPSA/pypsa-eur1 question · 1.2k stars Thu 3paper · Interconnection Queue Times by Region, 2019–20261 question Fri + 3 more staged, ranked the same way
06countaggregate

you › What have I been reading most since March?

aggregate(date_from="2026-03")

cursor › Tariffs and inflation is your biggest topic since March, with grid and data centers close behind:

tariffs & inflation96 grid & data centers84 interconnection queues53 GLP-1 trials41

Retrieval

Ask for what you've saved.

No new app, no chat UI. Opyt is a plain MCP server, so anything that speaks MCP to a local process can read your store: Claude Code, Claude Desktop, Cursor, Codex, Windsurf, etc. Your client queries it directly and does the reasoning on the subscription you already pay for. Everything lives in one local SQLite database in ~/.opyt.

07asksearch · open

you › Are tariff price effects one-off or recurring?

search("are tariff price effects one-off or recurring")

claude › Three of your own sources speak to this. The closest is a post you bookmarked yourself in June:

08readsitting

you › Read everything I've got on GLP-1 drugs and muscle loss, and lay out what it actually claims.

sitting(action="read", lens="claims")

claude › Eleven claims came back, each paired with what would break it and each citing the documents it came from. The first three:

claim, in publication orderwhat would disprove it
The headline number hides what came off: fat and muscle go together, and most trials report only the total. Body-composition scans published beside the headline figure.
A 28% loss is not comparable to a 15% one unless both measured lean mass the same way. One measurement standard adopted across the trials.
At population scale the binding constraint is people staying on the drug, not how well it works. Oral dosing flattening the discontinuation curve.
09keephopper

you › Save this: arxiv.org/abs/2608.01144

hopper("arxiv.org/abs/2608.01144")
routed → paper ingester preview: "Marginal Cost Allocation for Large Flexible Loads" · 22 pages · confirm?

you › yes

claude › Saved in full text, searchable beside everything else.

Saving something is one sentence. No folders, no tags, no deciding where it belongs.

Collaboration

Work from what other people have already read.

Reading from someone else's knowledge base is frictionless, so you can learn and build off each other's interests. One question reaches everyone who shared theirs with you.

10shareshare · accept · search

you › I need to pick an eval setup for retrieval. Has anyone already been through this?

search("evaluating retrieval quality", kb="ana")

search("evaluating retrieval quality", kb="ravi")

search("evaluating retrieval quality", kb="priya")

claude › Two of them have. Ana read the framework and its repo in March, and Ravi has the benchmark it is usually compared against. Nothing on this in yours:

Sharing sends the whole knowledge base to a copy Opyt hosts, and it stays readable until you take it back. unshare cuts off one reader, or every reader plus that copy. It goes one way: accepting gives you read access and gives them nothing, and what the service records is that you read, never what you searched for.

Cost

What it costs.

Opyt is free and MIT-licensed. The expensive part is the reading and reasoning, and that runs on the AI subscription you already pay for. One metered key covers the rest.

Your AI clientevery question, every answer, all the reasoningyour existing subscription
OpenRoutersorting and extracting what comes in on a cheap open model, plus the embeddings search ranks withpay-as-you-go credits
Xposts, timelines, profiles, threadsfree, through your own browser
Everything elsepapers, blogs, GitHub, Substack, any URL you hand itfree

Install

Install.

Pick your client.

  1. Run this in your terminal.
    # once per machine, if you don't already have uv curl -LsSf https://astral.sh/uv/install.sh | sh claude mcp add Opyt -- uvx --from opyt==0.1.0a4 opyt-mcp
  1. Restart your client. Opyt appears as twelve tools.
  2. Say onboard. A browser tab opens for the one key it needs, and you click Approve. Nothing is ever pasted into chat. It then looks for a browser already signed into x.com, which is how Opyt reads X.
  3. Say oracle to pick who to track, ranked from your own bookmarks and follows.

Star us on GitHub

Opyt's free and MIT-licensed. If it's useful to you, star it. That's how other people find it.

star on github
use casesnine questions it answers, with the tools that answer them
how Opyt workswhat each tool does, and where content comes from
Opyt vs the alternativesReadwise, Obsidian, NotebookLM, Grok and sixteen more
github.com/maimond123/Opytsource, issues, MIT license
hello@useopyt.comquestions and feedback
© 2026 Opyt Built with ❤️ in New Jersey MIT Use cases Compare Docs GitHub