# microsoft/TaskWeaver

The first "code-first" agent framework for seamlessly planning and executing data analytics tasks.

Repository: https://github.com/microsoft/TaskWeaver
Canonical: https://ross.abutalabs.com/products/taskweaver
Homepage: https://microsoft.github.io/TaskWeaver/
Language: Python
License: MIT
License Family: permissive
Topics: agent, llm, copilot, ai-agents, code-interpreter, data-analysis, openai
Archived: true
Last push: 2026-03-23T06:17:21+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 73, release rhythm 8, longevity 77
- inputs: {"age_days": 1087, "days_push": 163, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6172, forks 781 (observed 2026-08-28T04:09:36.913021+00:00)

## What it is
TaskWeaver is a code-first agent framework from Microsoft that plans and executes data analytics tasks by generating and running code snippets with plugin support. It preserves both chat history and code execution state, including in-memory data like pandas DataFrames, across a conversation.

## Use cases
- analyze csv files with natural language questions
- build an ai agent that writes and executes python code
- chat with tabular data using pandas dataframes
- create a data analytics copilot with custom plugins
- run stateful multi-step data analysis via llm
- serve a code interpreter agent locally with open llms

## When to choose
- you need an agent that executes code and keeps in-memory data state across turns
- you want a data analytics assistant with customizable plugins and domain examples
- you prefer self-hosted or containerized code execution with local llm support

## When to avoid
- you only need simple text-only chat without code execution
- you want a no-code agent builder rather than a python framework
- your workload requires non-python execution environments

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, data-science, plugin-system, chatbot
- domain: large-language-models, data-science
- platform: python, cross-platform, cli
- tags: code-interpreter, data-analytics, code-first-agent, pandas, stateful-conversation, openai, microsoft, ai-agents, data-engineering, natural-language-processing, docker

## Member repositories
- microsoft/TaskWeaver (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:36.913021+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:47:27.274913+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/TaskWeaver (fetched 2026-08-28T04:09:36.913021+00:00, sha cfc98c20aa36)
  - homepage: https://microsoft.github.io/TaskWeaver/ (fetched 2026-08-29T08:44:08.067689+00:00, sha 9299baacf305)
- Data as of 2026-08-30T08:39:29.467469+00:00.