Runprompt
Dev tools
command line prompt runner
SHOW HN2025
How HN took it
8 positive, 16 neutral or asking, 0 mixed, 1 critical, out of 25 comments about the project. Thread score +0.27.
Idea: 4 for, 0 againstExecution: 1 for, 1 againstDocs: 1 for, 0 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
Pipe-friendly, lightweight alternative to agent frameworks
From 2 comments
This is really clever. Dotprompt as a thin, pipe-friendly layer around LLMs feels way more ergonomic than spinning up a whole agent stack. The single-file + stdlib approach is a…
Barathkanna on HNThis is pretty cool. I like using snippets to run little scripts I have in the terminal (I use Alfred a lot on macOS). And right now I just manually do LLM requests in the…
cootsnuck on HN
Single-file stdlib-only design
From 1 comment
This is really clever. Dotprompt as a thin, pipe-friendly layer around LLMs feels way more ergonomic than spinning up a whole agent stack. The single-file + stdlib approach is a…
Barathkanna on HN
Objections
Invalid LLM output passes silently down the chain
From 1 comment
If the LLM returns invalid schema the next link in the chain will send that through since it defaults to string input if the JSON doesn't parse. Maybe I should make it error out…
chr15m on HN
Requests
Support for local LLMs via custom base URL (Ollama)
From 1 comment
Can the base URL be overridden so I can point it at eg Ollama or any other OpenAI compatible endpoint? I’d love to use this with local LLMs, for the speed and privacy boost.
dymk on HN
Response caching for iterating on pipelines
From 1 comment
It would be cool if there were some cache (invalidated by hand, potentially distributed across many users) so we could get consistent results while iterating on the later stages…
__MatrixMan__ on HN
Shebang support in prompt files
From 1 comment
Can it be made to be directly executable with a shebang line?
cedws on HN
Questions
How robust is JSON schema enforcement when chaining steps
From 1 comment
This is really clever. Dotprompt as a thin, pipe-friendly layer around LLMs feels way more ergonomic than spinning up a whole agent stack. The single-file + stdlib approach is a…
Barathkanna on HN
Why not use existing markdown files or simonw's llm tool
From 2 comments
Why this over md files I already make and can be read by any agent CLI ( Claude, Gemini, codex, etc)?
orliesaurus on HNPerhaps instead of writing an llm abstraction layer, you could use a lightweight one, such as @simonw's llm.
benatkin on HN
Alternatives named
Simon Willison's llm CLI
From 2 comments
Perhaps instead of writing an llm abstraction layer, you could use a lightweight one, such as @simonw's llm.
benatkin on HNThats pretty good, now lets see simonw's one...
swah on HN
Microsoft Prompty format
From 1 comment
Interesting! Seems there is a very similar format by Microsoft called `.prompty`. Maybe I'll work on a PR to support either `.prompt` or `.prompty` files.…
oddrationale on HN
Google dotprompt Python library
From 1 comment
Looks like Google has packaged dotprompt into a Python library, might allow you to make the codebase leaner: https://github.com/google/dotprompt/tree/main/python/dotpromptz I…
threecheese on HN