WUPHF
Productivity
agent knowledge wiki
SHOW HN2026
How HN took it
10 positive, 26 neutral or asking, 2 mixed, 10 critical, out of 48 comments about the project. Thread score -0.01.
Idea: 4 for, 6 againstDesign: 3 for, 1 againstDocs: 0 for, 2 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
BM25-first over vector DBs is a good call
From 1 comment
love the bm25-first call over vector dbs. most teams jump to vectors before measuring anything
Unsponsoredio on HN
Objections
LLM-maintained markdown collections may degrade quality without human curation
From 1 comment
The few scientific studies out there actually show a degradation of output quality when these markdown collections are fully LLM maintained (opposed to an increase when they’re…
stingraycharles on HN
Name and product seem like satire or redundant
From 2 comments
I thought this was parody at first as well for a redundant useless product as it was named after the redundant useless product of the same name from The Office (Wuphf.com)
arikrahman on HNI no longer can tell if this is satire or not: https://wuphf.team/ It looks like gastown. Using AI is like children playing with a dollhouse. If it is satire, well done.
410298 on HN
Agent-generated content gets confidently wrong over time
From 2 comments
The "garbage facts in, garbage briefs out" caveat is the part I'd want stress tested. In my own LLM features the context that decays fastest is what agents wrote without a human…
Abby_101 on HNOne practical approach that works is separating the capture layer from the promotion layer. Agents can draft freely, but anything that gets promoted to trusted status needs a…
ryanshrott on HN
Requests
Configurable wiki directory
From 1 comment
Couldn't you instruct your LLM to make the starting dir configurable?
hansmayer on HN
Support for arbitrary OpenAI-compatible endpoints like DeepSeek
From 1 comment
Would be great if the provider layer supported arbitrary OpenAI-compatible endpoints — DeepSeek, for example. Any plans?
LanMeng-LM on HN
Questions
Does it support multiple humans?
From 1 comment
I couldn't find out on the fly: does it support multiple humans? A lot of these tools (paperclip) don't, but for our small, agent-supported team, that would be the feature we're…
psteinweber on HN
Does promotion require human review or can agents self-promote?
From 1 comment
The "garbage facts in, garbage briefs out" caveat is the part I'd want stress tested. In my own LLM features the context that decays fastest is what agents wrote without a human…
Abby_101 on HN
How to deploy securely for sensitive documents?
From 1 comment
How do you anticipate teams deploying this? I’m wary of GitHub for sensitive business documents, and wonder what an easy secure agent friendly deployment looks like. Cloudflare…
sails on HN
Alternatives named
Obsidian vault with a plugin
From 1 comment
why not an Obsidian vault with a plugin?
davedigerati on HN
Other LLM wiki projects such as commonplace and Shoggoth.db
From 2 comments
Reviewed: https://zby.github.io/commonplace/agent-memory-systems/reviews/wuphf/ It is a third llm wiki on front page in 24 hours! Obviously it is a hot topic. I have my own horse…
zby on HNThe space of self building artefacts is interesting and is booming now because recent LLM versions are becoming good at it fast (in particular if they are of the "coding" kind).…
GistNoesis on HN