DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to use Smol: what you can do with it

What is Smol? smol is a deliberately tiny agent loop in Python, with golfed versions in Go, Clojure and PHP, that talks to any OpenAI Responses API compatible endpoint and executes model-requested shell commands.

How do I use it? Clone the repo and read Smol.py; then run a small local proxy that holds your API key or session and exposes a Responses API endpoint; run python3 smol.py http://127.0.0.1:8787/v1/responses. Full steps below.

Trended earlier

View repo on GitHub

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer

What do you want to do?

3 use cases

  • Understand a full agent loop in one afternoon

    Run Smol, a deliberately tiny Python agent with one shell tool, against any Responses API endpoint, then read its code with the README's questions.

    View prompt
    Read the code of smol.py from smol-env/smol and answer all questions from the README: what it does, why it has one tool, how compaction keeps a stable prefix and what that does to caching and cost, and its failure modes. Then explain it line by line.

  • Compare a minimal agent loop in four languages

    Run the golfed Smol in Go and read the Python, Clojure and PHP versions to see the essential agent loop with nothing extra.

    View prompt
    Compare golf/smol.py, golf/smol.go, golf/smol.clj and golf/smol.php from smol-env/smol in a table: lines of code, dependencies, how each handles the tool call and context. Then port it to <language> using only the standard library.

  • Use Smol day to day with line editing and history

    Wrap Smol in rlwrap for arrow key editing and a saved prompt history, and run it in a sandboxed environment that has access to your endpoint.

    View prompt
    Write me a minimal standard-library-only proxy that keeps my API key outside smol and forwards requests to <provider> on 127.0.0.1:8787, then a docker run command that starts smol with rlwrap history inside a throwaway container.

Quick start with Smol

  1. Check you have what it needs: Python 3 or Go; An OpenAI Responses API compatible endpoint, such as a local proxy that adds your key.
  2. Open smol-env/smol on GitHub and read the README and license.
  3. Clone the repo and read Smol.py
  4. Run a small local proxy that holds your API key or session and exposes a Responses API endpoint
  5. Run python3 smol.py http://127.0.0.1:8787/v1/responses
  6. Type a prompt at the > prompt; empty input exits

These steps come from our top use case, Understand a full agent loop in one afternoon, which also has a copy-paste agent prompt.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for smol-env/smol

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Smol alternatives · Best AI repos for Software and DevOps

About this repo

  • Who it is forDevelopers who want to understand or build on an agent loop small enough to read in an afternoon.
  • Why it is hot137 stars, launched 86 days ago
  • LanguagePython
  • Sourcegithub.com/smol-env/smol · Back to Hot repos

Requirements

  • Python 3 or Go
  • An OpenAI Responses API compatible endpoint, such as a local proxy that adds your key

Add this badge to your README

Maintain Smol? Show readers where to find these use cases. Use cases on Deduped

Markdown 

[![Use cases on Deduped](https://dedupedai.com/badge/smol-env/smol.svg)](https://dedupedai.com/use-cases/smol-env-smol/?utm_source=github&utm_medium=badge&utm_campaign=smol-env-smol)

HTML 

<a href="https://dedupedai.com/use-cases/smol-env-smol/?utm_source=github&amp;utm_medium=badge&amp;utm_campaign=smol-env-smol"><img src="https://dedupedai.com/badge/smol-env/smol.svg" alt="Use cases on Deduped" height="20"></a>

A static image, no tracking script. About the badge

Questions about Smol

What is Smol?

smol is a deliberately tiny agent loop in Python, with golfed versions in Go, Clojure and PHP, that talks to any OpenAI Responses API compatible endpoint and executes model-requested shell commands. It is an open-source project on GitHub: https://github.com/smol-env/smol.

How do I use Smol?

Clone the repo and read Smol.py; then run a small local proxy that holds your API key or session and exposes a Responses API endpoint; run python3 smol.py http://127.0.0.1:8787/v1/responses. The use cases on this page give the full steps and a copy-paste agent prompt for each job.

Who is Smol for?

Developers who want to understand or build on an agent loop small enough to read in an afternoon.

What can I do with Smol?

3 use cases so far, for example: Understand a full agent loop in one afternoon; Compare a minimal agent loop in four languages; Use Smol day to day with line editing and history. Each has steps and a copy-paste agent prompt.

What do I need to run Smol?

Python 3 or Go; An OpenAI Responses API compatible endpoint, such as a local proxy that adds your key.

What are alternatives to Smol?

Other hot AI repos in Software and DevOps: Deep Infer, Harness2, Orca, Opencode, Graphify.

Is Smol safe to run?

Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.