A small web app I threw together while working in an IT service desk role as a personal knowledge base. Sharing it here in case it’s useful to others in similar roles, or anywhere else this kind of thing might help.

A lot of orgs have shared knowledge bases, but in practice they’re often outdated, incomplete, or hard to search. What I noticed instead was a lot of knowledge silo's; everyone keeping their own notes in Word docs, text files bookmarks, etc.

This was basically built to act as my own personal Google. I could curate a corpus of the issues I ran into most often and quickly reference them while handling tickets.

I’ll admit it’s not the most efficient or polished approach, partly because it was designed around a few technical limitations at the time, but it ended up being genuinely useful for day-to-day support work.

Hopefully some of you will be able to find this useful but also would be interested hearing any feedback.

Comments URL: https://news.ycombinator.com/item?id=48181397

Points: 2

# Comments: 0

It got bigger and it became a pain to dig through directories of markdown so then I asked CC to build a python script to convert the md folder into browsable html. I found it really useful in terms of both being able to browse the html and also being able to add to it and interrogate it via agent.

Then I decided to start using it for other topics/projects so I tried creating a seed prompt that would recreate my whole system from a single prompt. And it works and I'm finding it great to use for those projects too. It's totally flexible and you change whatever you want about it by just telling it.

Also it's really cool that we can now build our own personal easily customisable SaaS from a single prompt.

*Please note* that the prompt instructs your agent to build a python script that creates a static html site from the markdown. It does ask permission first (well it's supposed to). Look at Phase 3 in the markdown file to see these instructions; it's only 10 lines of prompt and easy to follow. Don't trust unknown prompts without reading them!!!

I've only tried this with Opus 4.7 but it should work just as well with 5.5 and I'm sure other models too. I've not tried it with OpenClaw yet but will do soon. To make the most of it in OpenClaw the seed prompt would probably need to change a bit.

I don't recommend installing directly this into an existing project folder. Please only install in a fresh directory, I have no idea what it would do if there were existing files in the folder but you might not be happy about it. It would probably be sensible but you never know.

If you want it to access a folder with a coding project, clone your repo there later. You don't want to have your main coding agent get it's context memory polluted by managing this system.

To install simply paste the prompt into a new empty cowork project or an empty directory using Claude Code. Then let it onboard you. If you don't like the onboarding path then tell it what you want and it should hopefully adapt itself.

Anyway thanks for reading; hope it's useful or interesting.

Comments URL: https://news.ycombinator.com/item?id=48154710

Points: 2

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=48061548

Points: 3

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=48035324

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47933743

Points: 1

# Comments: 0

Time and time again, I've seen teams try to change something to a better way, only to realize why it was done the "worse" way in the first place. Documenting decisions not taken is just really hard, ADRs help but doing it for everything is just too noisy and time-consuming. What's more, most of the decisions just detail what was decided, which is the easy part.

I've been wanting to change this for a long time, and with LLMs, we now can. LLMs are diligent about documenting, and all you need is an instruction in AGENTS.md. That's why I built gnosis.

Simply tell your agent to run it after planning and when done, and gnosis provides the agent with all direction necessary.

Gnosis tells the agent to document only information it got from the human, not anything it can find by itself, and especially to document why alternatives were rejected. It uses a JSONL append-only log and a SQLite index (for retrieval), making it simple, fast, and convenient.

I'd appreciate it if you tried it out and gave me feedback!

Comments URL: https://news.ycombinator.com/item?id=47920641

Points: 6

# Comments: 4

Comments URL: https://news.ycombinator.com/item?id=47901737

Points: 4

# Comments: 4

- Rebuilt the iOS app with an Android app on the way

- expanded both the MCP and internal agent chat toolkit immensely

- A custom, CodeMirror6-based markdown editor with obsidian-style rendering

- A dashboard view that provides a daily summary of atoms created or updated in the last day

And many bug fixes and improvements across the board. Atomic is MIT licensed. You can download the desktop app, but the true power is unlocked by self hosting an atomic server, which any client (web, mobile, or desktop) can connect to from anywhere. You can add content to your knowledge base directly, or via RSS feed, web clipper, mobile share capture, obsidian sync, or REST api.

Comments URL: https://news.ycombinator.com/item?id=47889110

Points: 62

# Comments: 43

Comments URL: https://news.ycombinator.com/item?id=47863280

Points: 2

# Comments: 0

There’s a truckload of tribal knowledge we’ve accumulated; major decisions, gotchas, user feedback driven changes. Providing this to our agents manually every time is very mundane.

We built Kilroy to solve this in a simple way: we let our agents leave notes for each other. This allowed us to keep the form factor minimal: markdown posts with linear comments. Under the hood it’s Postgres + an auth (better-auth) + an MCP + a small web UI (React). We ship Claude Code and Codex plugins that bundle the MCP + a skill.md that teaches the model when to read and write posts.

We designed Kilroy to be autonomous. The same way agents today run a typechecker after a patch autonomously. The combination we found to work best for us was: make agents write prolifically, expose a search interface designed for agents to quickly decide if a post is relevant, and expose a binary switch to purge stale posts.

Would love to get feedback!

Comments URL: https://news.ycombinator.com/item?id=47791559

Points: 5

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47772114

Points: 1

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=47764322

Points: 3

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=47753554

Points: 2

# Comments: 1

Github repo: https://github.com/VectifyAI/OpenKB

Comments URL: https://news.ycombinator.com/item?id=47752392

Points: 6

# Comments: 2

Comments URL: https://news.ycombinator.com/item?id=47709940

Points: 2

# Comments: 5

Comments URL: https://news.ycombinator.com/item?id=47706592

Points: 3

# Comments: 1

Comments URL: https://news.ycombinator.com/item?id=47698966

Points: 3

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47674619

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47672436

Points: 1

# Comments: 1

During my daily working life, I have tons of office documents with knowledge from all teams, and as an IT Architect, I need to combine them altogether to handle complex deep research (which normal LLM definitely could not help). That is the originally reason I built DocMason, and I am using it in everyday which support me on lots of complex topics.

I have already open-sourced this repo. And I think it takes Karpathy's concept a step further for real-world usage in three ways: 1. It could handle most kinds of office docs (pptx, docx, excels, even .eml). And really extract multimodal information from all IT architecture diagram or excel sheets. 2. It is running as a Real APP but not a naive RAG tool. DocMason could run smoothly and intelligently to prepare environment, auto update, and auto incrementally sync Knowledge base. 3. Most importantly it is running in Native AI Agents, which could leverage powerful AI Agents engine (e.g. Codex or Claude Code)

View detail architecture diagram in DocMason Readme, and then download have a try :) You will find it could help a lot during daily work. Would love to hear your feedback and issues in Github!

Comments URL: https://news.ycombinator.com/item?id=47640770

Points: 11

# Comments: 0