Reduce Code intelligence etc

Is there a way to reduce the amount of Code intelligence available to students in an edu repl/project?

As more Code intelligence features and ghostwriter prompts are added to replit, these can be more of a confusing distraction than help for students.

Is there a template or instructions on how to customise the features that can be forked? are code intelligence features defined in the .replit and replit.nix or other files??

ideally I would like to strip it down to simple autocomple and some simple highlighting of basic syntax errors.



Hi @CCR010 , welcome to the forums!
Replit is now using pyright-extended from the original simple Code Intelligence.
You can completely disable CI by going to Tools > Settings > Code Intelligence and toggle it off.
Hope this helps!


Pyright-extended is configurable with the pyproject.toml file or with a pyrightconfig.json file.


Can anyone provide me with a bare bones project with a commented config file with options that I can turn on and off and experiment with? Thank you. :slight_smile:

I found a file called .replit when I toggled on “Show Hidden Files”. I messed around in this file, which appears to have various code intelligence settings, but it didn’t seem to change anything. (I have no idea what I’m doing though…)

For reference:
In the newest python templates, it is very clear how to configure Code intelligence.
So create a new python repl and open the pyproject.toml file.

Most code intelligence is controlled by ruff.
The list of enabled categories of rules is in the select variable: add or remove rulesets here.
You can disable individual rules in the ignore variable.
The categories and specific rules are found at Rules - Ruff.

More settings such as line length can be configured by putting more variables in the file, sometimes in separate sections.
Details are found at Settings - Ruff.

Example pyproject.toml file:

name = "python-template"
version = "0.1.0"
description = ""
authors = ["Your Name <>"]

python = ">=3.10.0"

useLibraryCodeForTypes = true
exclude = [".cache"]

select = ['E', 'W', 'F', 'I', 'B', 'C4', 'ARG', 'SIM', 'PIE']
ignore = ['W292', 'W293', 'SIM105']

line-length = 90

  combine-as-imports = true

requires = ["poetry-core>=1.0.0"]
build-backend = "poetry.core.masonry.api"

Invalid configuration such as putting a nonexistent ruleset or specific rule, or a typo in a variable name may break code intelligence, disabling it. Be sure that code intelligence is working when modifying configuration (it takes a few seconds for changes to take effect).
Also, code intelligence cannot be configured with the CLI (shell).

1 Like