start here

start using Python with your agents

sign up, open a project, and add Python to its setup script in settings → server.

Start 7-day trial with Python
integrationstechnologyupdated
schematic diagram of the work aldena agents do through Python
Python

python ai agent: agents that write the script and then run it

add python to a project's setup script and its agents get python3, pip3 and venv through the bash tool, so they build environments and run the code they write.

runs on
conversation sandbox
installed by
setup script
category
runtime

python on a project is an install in each conversation's own sandbox, not a sandbox that evaporates between messages. once the setup script has run, python3, python, pip3 and the venv module are on the path and every agent in the project reaches them through the bash tool.

what agents do with it

the point of having python in the conversation's sandbox is that the agent that writes a script can run it against the real files. it creates .venv, installs from requirements.txt, runs the script, reads the traceback, edits the code, and runs it again. a data file the agent downloads lands on the same machine, so the agent can load it without anyone moving bytes around by hand.

the matching skill

python ships with a skill of the same name, carrying the guidance on virtual environments, pip3 install -r requirements.txt, and running scripts. skills are versioned documents that load only for the agents you assign them to, so a build agent carries it and the intake agent in the same project does not.

example tasks

  • 01

    load the export into a dataframe and tell me which rows are malformed.

    the agent builds a virtual environment, installs what the script needs, runs it against the file in the conversation's sandbox, and reports the real row numbers.

    bashwriteread
  • 02

    the scraper is failing on the new page layout. fix it.

    it runs the script to see the actual traceback, edits the selector, reruns until the output is right, and shows you the diff before anything is pushed.

    basheditgithub_git_push

what to know before installing

the project needs a sandbox. boat's ubuntu image already ships python3; on any provider, the setup script installs pip, venv and whatever else the code expects. a new conversation gets its own sandbox, and the setup script runs on it again.

how to install Python

Python goes in the project's setup script, not once per team and not on your laptop.

  1. give the project a sandbox

    Python runs in each conversation's sandbox, on your team's own sandbox provider account. an owner or admin connects a provider once on the team's providers page, then picks it and a default size for the project, at creation or later in settings → server. a project with no sandbox has nowhere to run it.

  2. add it to the setup script

    settings → server holds the project's setup script, a bash script that runs once on each new sandbox, in /project, with sudo. add Python's install commands there and save.

  3. the next sandbox runs it

    the script runs when a conversation's sandbox is created or reused, on the conversation's first turn. settings shows the last run with its exit code and output, and a failed run leaves the sandbox usable rather than blocking the project.

  4. agents use it through bash

    from then on every agent in the project reaches Python with the bash tool, on the same filesystem as the checkout they are working in. the conversation's sandbox keeps it between turns, sleeping when idle and waking with the next one.

where Python actually lives

each conversation in a project gets its own sandbox on your team's sandbox provider account, and the project's setup script in settings → server decides what is on it. two projects never share a filesystem, so a version pinned for one client cannot collide with another's.

the sandbox provider bills its machine time to your team's account while it runs, and aldena adds nothing. the sizes are on isolated projects, and what agents may run in the sandbox is governed by the project's tool permissions.

faq

what does installing python on a project give the agents?

python3, python, pip3 and the venv module in each conversation's sandbox, reachable through the bash tool. agents create an isolated environment, install from requirements.txt, and run the script on the same machine that holds the checkout and the data.

will agents install packages into the system python?

the shipped python skill tells them to create a virtual environment with python3 -m venv .venv and activate it before installing, so project dependencies stay in the project. the skill loads only for the agents you assign it to.

can an agent run a long data job in a project?

yes, in the conversation's own sandbox, at the size an owner or admin picked for the project (on boat, from 2 vcpu and 4 gb of ram on small up to 16 vcpu and 32 gb on x-large). the sandbox provider bills the machine time to your team's account.

does installing python in one project affect another?

no. each project has its own setup script, each conversation gets its own sandbox, and two projects never share a filesystem. one project can pin an old dependency set for a client while another runs something current, with no interference between them.

ready when you are

spin up your first project.

one project per client, product, or piece of work, staffed with intake, requirements, build and code review agents.