Oiya

Oiya builds structured institutional memory for AI agents. We also make OpenLore, the open-source knowledge server for shared agent context.

[Oiya @Oiya_ai]

If you've been looking for knowledge that works across tools and teams, or a practical way to do context engineering, this is what OpenLore is for. Self-hosted. Open source.

GitHub: https://t.co/LJFszzoNml
Website: https://t.co/VwkghWMQpS

[Alfie Lambert @AlfieOiya]

We just hit our first 1k stars on GitHub ⭐️

OpenLore started from solving a problem we faced every day (like all good ideas): once multiple AI agents share knowledge, prompts and folders stop being enough.

You need permissions, provenance and rules the agents can’t ignore.

Nice to know we’re not the only ones thinking about this.

If you’re not using OpenLore, there are now 1,000 people whose agents are better set up than yours 👀

https://t.co/bOGblxvL5A

[Arindam Majumder 𝕏 @Arindam_1729]

Spent the weekend at IoTricity, and it was a blast.

Hackathons are one of the best places to learn. You meet developers, pick their brains, and see how people actually build under pressure.

I got to give a talk on what I've learned about shipping at hackathons and how a team can work together to build something cool.

The main point:

→ Using agents is the new norm
→ Sharing context across your team matters more than ever
→ How you share it (the process) decides how fast you ship

I also showed how OpenLore (https://t.co/v3Gki4uUV3) by @Oiya_ai makes this whole flow way smoother.

I've made the slides public if you want to check them out: https://t.co/5J9EwKNg49

The feedback was great, and the team's response was amazing. Looking forward to more events like this 🙌

[Oiya @Oiya_ai]

When multiple AI agents (and humans!) can write to the same knowledge, “who changed this?” becomes a real infrastructure question.OpenLore records the agent, the human behind it, how identity was verified, and what changed.

https://t.co/Eqp895WoA9

[Oiya @Oiya_ai]

Your CLAUDE.md says “keep this file under 800 lines.” Your agent hits 1,212 and decides it “needed the extra space”.

That’s the problem with putting governance in prompts. OpenLore enforces the rule at the write path.
https://t.co/Z1Bbi15iJ6

[Oiya @Oiya_ai]

As files grow, every future task gets more expensive and more cluttered.

OpenLore lets you enforce context budgets, so your agents can’t turn useful knowledge into a 30,000-token dumping ground.

https://t.co/e8EgDGqE8b

[Oiya @Oiya_ai]

Context engineering doesn't just mean writing a better system prompt.

It means deciding:

what should an agent know?
when should it know it?
where should that knowledge live?
who can change it?
how do you stop it getting worse over time?

It's an infrastructure problem.

[Oiya @Oiya_ai]

Your agents are consuming context every time they work, but do you actually know which files they’re reading and where the tokens are going?

OpenLore v0.7.0 adds analytics for shared agent knowledge.

See what agents read. Then make the context better.

https://t.co/WBVZohPax4

[Mr. Ånand @Astrodevil_]

This open-source project makes crucial context shareable across your team as knowledge base.🧠

Most personal agents work in isolation in a remote team setup. They forget context, duplicate knowledge, or rely on stale docs.

Exploring it myself today!

OpenLore solves this by creating one shared knowledge base that every agent can access.

Your docs stay as regular Markdown. Agents access them through SSH or MCP using commands they already know → ls, cat, grep, find.

No vector database. No ingestion pipeline. No LLM required.

✅ What makes it useful:

→ One source of truth for all agents
→ SSH + MCP transport
→ Identity-based access control
→ Live updates, no duplicates
→ Governed writes with validation
→ Embed docs into binaries
→ Web view for humans

Works for team memory, API docs, runbooks, agent skills, and artifact repos.

Instead of copying Markdown everywhere or letting agents drift out of sync, you serve knowledge from one place.

Your knowledge. Your control. Team agents stay aligned.

[Alfie Lambert @AlfieOiya]

Want to see which docs are getting bloated, what your agents are actually reading, and where context is wasting your tokens?

OpenLore v0.7.0 shipped last week.

OpenLore can surface the problem, then your agent can fix the structure.

Same knowledge. Less context. Better signal.

Free & open source: https://t.co/bOGblxvL5A

[Adil Karim @_aakarim]

Your team’s docs shouldn’t live in Git. They should live in a remote filesystem your agents can search.
Launched OpenLore at the Claude Code Copenhagen meetup last week. It serves your docs as a shell over SSH. One Go binary. https://t.co/Rbv3oDHe6o

[Alfie Lambert @AlfieOiya]

My X bookmarks used to be a graveyard.

Now they automatically feed my second brain and become context for my agents. Feed this to your agent and clean up your own bookmark graveyard! https://t.co/a0CbZ4BE0o

[Alfie Lambert @AlfieOiya]

https://t.co/LNumiLx5qz