posit-dev / posit-dev/btw

a tool to explore/manipulate data frames

Open
#52 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
R
Stars
137
Forks
10
Avg merge
10h 48m
Merged PRs (30d)
4

Description

I'm curious whether we can implement a tool that would safely allow models to explore data frames more dynamically. I've been feeling that the "env" subset of tools is a bit lacking compared to the others and I think this would notch in nicely.

One of the tools from querychat is a good example of this; read-only access using only SELECT SQL statements. I think it'd be nice if the models could write in dplyr/tidyr instead, here—with the hunch that the tool will be easier to deploy if it's "all R" and easier to read by common users than if it were in SQL—though I'm short on ideas of how we could do so without opening up the "run arbitrary R code" path.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the env tool area in this repository and the referenced querychat.R example, especially its read-only SELECT tool. Define what safe dplyr/tidyr data-frame exploration should support and how completion can be verified without permitting arbitrary R code.

Written by the indexing model from the issue text.

Assessment

Tech stack
r
Domain
data, tooling
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.