Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL • Hacker News Board

Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#1

Post by calebhwin » Thu, Feb 12, 2026, 5:42 PM UTC

Hi HN,

Been hacking on a simple way to run agents entirely inside of a Postgres database, "an agent per row".

Things you could build with this: * Your own agent orchestrator * A personal assistant with time travel * (more things I can't think of yet)

Not quite there yet but thought I'd share it in its current state.

Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL
github.com

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#2

Post by tibbar » Thu, Feb 12, 2026, 6:58 PM UTC

This is mind-bending. I can't imagine that it performs well enough to be particularly fit for production just yet but..... wow.

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#3

Post by readme » Thu, Feb 12, 2026, 7:01 PM UTC

i love postgres and pgvector... this is exactly to my tastes

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#4

Post by calebhwin » Thu, Feb 12, 2026, 7:07 PM UTC

readme wrote: post #3

i love postgres and pgvector... this is exactly to my tastes

Same, it's in a similar vein as pgvector - probably not as performant as turbopuffer or chroma but you get the benefits of being in a really nice ecosystem.

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#5

Post by wwweston » Thu, Feb 12, 2026, 7:09 PM UTC

This... does not seem like separation of concerns.

Not to mention that the data layer seems like the one where you want to keep things most deterministic.

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#6

Post by tibbar » Thu, Feb 12, 2026, 7:10 PM UTC

calebhwin wrote: post #4

readme wrote: post #3

i love postgres and pgvector... this is exactly to my tastes

Same, it's in a similar vein as pgvector - probably not as performant as turbopuffer or chroma but you get the benefits of being in a really nice ecosystem.

Is it? pgvector is primarily about indexing, I think? This feels more similar =AI in Google Sheets.

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#7

Post by danr4 » Thu, Feb 12, 2026, 7:17 PM UTC

this is fucking awesome

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#8

Post by xnx » Thu, Feb 12, 2026, 7:23 PM UTC

It feels like we're a week away from the Claw hype supplanting AI hype. Companies will start renaming things ClawX to get on the hype bandwagon.

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#9

Post by calebhwin » Thu, Feb 12, 2026, 7:26 PM UTC

tibbar wrote: post #6

calebhwin wrote: post #4

Earlier quoted context omitted.

Same, it's in a similar vein as pgvector - probably not as performant as turbopuffer or chroma but you get the benefits of being in a really nice ecosystem.

Is it? pgvector is primarily about indexing, I think? This feels more similar =AI in Google Sheets.

True though pgvector is not deterministic in a strict sense since HNSW is probabilistic.

Re: Show HN: Pgclaw – A "Clawdbot" in every row with 400 lines of Postgres SQL

#10

Post by calebhwin » Thu, Feb 12, 2026, 7:28 PM UTC

xnx wrote: post #8

It feels like we're a week away from the Claw hype supplanting AI hype. Companies will start renaming things ClawX to get on the hype bandwagon.

It's just an abstraction people are excited about at the moment. Langchain was an exciting abstraction at one point.

My bet is we converge on a super minimal modelcomputer architecture.