# AI use cases for Agentenv Framework (scaleapi/agentenv-framework) Repo: [scaleapi/agentenv-framework](https://github.com/scaleapi/agentenv-framework) | Use cases: 4 | Page: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/ AgentEnv Framework is a Python SDK and CLI for building, deploying and running containerized agentic environments and the tasks that grade agents inside them. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Run the hello task to check your install: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/run-hello-task-to-verify-install/ - Run local environments in the explorer: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/run-local-environments-in-explorer/ - Build a custom environment with the SDK: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/build-custom-environment-with-sdk/ - Set up a clone for a contribution with a coding agent: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/contribute-change-with-coding-agent/ ## How to run the hello task to check your install with Agentenv Framework Repo: [scaleapi/agentenv-framework](https://github.com/scaleapi/agentenv-framework) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README Confirm agent-env works on your machine with the built-in hello task. **Who it is for:** A developer evaluating AgentEnv Framework who wants a quick check before building anything. **The problem:** Before investing in environments and tasks, you need to know the CLI installs and runs on your machine. **The outcome:** The built-in hello task runs and you know which version of agent-env is installed. ### Steps 1. Install uv from https://docs.astral.sh/uv/ if it is not already on your machine. 2. Run `uv tool install agentenv-framework`. 3. Run `agent-env run hello`. 4. Run `agent-env --version` to confirm the installed version. ### Prompt to paste into your agent ```text In the scaleapi/agentenv-framework repo, run `agent-env run hello` from , then run `agent-env --version`. Report the output of both commands and whether the hello task completed. Verify that the version printed matches the version you installed. ``` ### Caveat The hello task needs no Docker, model or configuration, but running environments you build yourself does need a Docker daemon. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/run-hello-task-to-verify-install/ ## How to run local environments in the explorer with Agentenv Framework Repo: [scaleapi/agentenv-framework](https://github.com/scaleapi/agentenv-framework) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Start the explorer on your own Docker daemon using the all-local config. **Who it is for:** An RL researcher or engineer who wants to build and inspect environments on a laptop or workstation before sharing them. **The problem:** You need a local setup where environments and agents run as containers, with a visual explorer to look at them. **The outcome:** The explorer starts with the all-local configuration, and environments and agents run as containers on your Docker daemon. ### Steps 1. Make sure a Docker daemon is running on your machine. 2. Copy `.agentenv/config.example.toml` to `.agentenv/config.toml` in your project folder. (inferred, check the README) 3. Run `uv tool install 'agentenv-framework[explorer]'`. 4. Run `agent-env up` from your project folder or a folder below it. ### Prompt to paste into your agent ```text In the scaleapi/agentenv-framework repo, prepare with an .agentenv/config.toml copied from .agentenv/config.example.toml, then run `agent-env up`. Report the address or message the explorer prints and whether it started. Verify that the Docker daemon is reachable before you start. ``` ### Caveat The explorer needs the explorer extra and a config file in the current folder or above it; an empty config file keeps every local default. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/run-local-environments-in-explorer/ ## How to build a custom environment with the SDK with Agentenv Framework Repo: [scaleapi/agentenv-framework](https://github.com/scaleapi/agentenv-framework) | Difficulty: Advanced | Time: 1 hour | Basis: From the README Add the SDK to a Python project and build an environment server for your tasks. **Who it is for:** A platform engineer or researcher who needs an environment specific to their tasks and wants to build it in Python. **The problem:** Standard tasks do not match your domain, so you need your own environment that agents can run in and be graded in. **The outcome:** Your Python project depends on agent-env and imports agent_env, and you have a starting point for an environment server that speaks the protocol. ### Steps 1. Run `uv add agentenv-framework` in your project folder. 2. Read the environment server SDK section in packages/agentenv-protocol/README.md. 3. Write your environment server using the SDK classes described there. 4. Build and run it with the agent-env CLI as described in the documentation at https://www.agentenvframework.com/docs. ### Prompt to paste into your agent ```text In , add agentenv-framework with `uv add agentenv-framework`, then follow the environment server SDK section in packages/agentenv-protocol/README.md of scaleapi/agentenv-framework to create a minimal environment server. Report the files you created and the commands you ran. Verify that `python -c "import agent_env"` runs without error inside the project environment. ``` ### Caveat The README names the environment server SDK and documents it in packages/agentenv-protocol/README.md, but the exact class names and build steps are in those docs, not in this summary. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/build-custom-environment-with-sdk/ ## How to set up a clone for a contribution with a coding agent with Agentenv Framework Repo: [scaleapi/agentenv-framework](https://github.com/scaleapi/agentenv-framework) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Clone the repo, sync dependencies and follow AGENTS.md before changing code. **Who it is for:** A contributor or a coding agent asked to fix a bug or add a small feature in agent-env. **The problem:** Changes must follow the repo map, test tiers and pull request rules, and larger changes need an issue first. **The outcome:** A working development clone with dependencies installed, and a plan that follows CONTRIBUTING.md and AGENTS.md. ### Steps 1. Run `git clone https://github.com/scaleapi/agentenv-framework && cd agentenv-framework`. 2. Run `uv sync --extra dev`. 3. Run `source .venv/bin/activate`. 4. Read AGENTS.md and CONTRIBUTING.md, and open an issue first if the change is larger than a bug fix. ### Prompt to paste into your agent ```text Clone https://github.com/scaleapi/agentenv-framework, run `uv sync --extra dev`, then read AGENTS.md and CONTRIBUTING.md. Report the test tiers and pull request title format you found, and the change you plan for . Verify that the virtual environment activates before you edit any code. ``` ### Caveat Pull requests need an approving review from a code owner and green CI, and a change that breaks plugin classes or entry points fails CI unless its title is marked with `!`. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/scaleapi-agentenv-framework/contribute-change-with-coding-agent/