# microsoft/promptflow

Build high-quality LLM apps - from prototyping, testing to production deployment and monitoring.

Repository: https://github.com/microsoft/promptflow
Canonical: https://ross.abutalabs.com/products/promptflow
Homepage: https://microsoft.github.io/promptflow/
Language: Python
License: MIT
License Family: permissive
Topics: ai, llm, chatgpt, gpt, prompt, prompt-engineering, ai-application-development, ai-applications
Last push: 2026-08-26T22:27:46+00:00

## Health v2 (maintenance only)
Score: 75/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 40, longevity 82
- inputs: {"age_days": 1160, "days_push": 7, "days_rel": 601, "gap_med": 8, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 11229, forks 1118 (observed 2026-08-28T04:10:46.510322+00:00)

## What it is
Prompt flow is a suite of Python SDK, CLI, and VS Code extension tools for the end-to-end development of LLM-based AI applications, linking LLMs, prompts, and Python code into executable flows. It supports prototyping, debugging with tracing, evaluation, CI/CD integration, deployment, and monitoring. Note: feature development ended April 2026 and the project retires April 2027 in favor of Microsoft Agent Framework.

## Use cases
- build and iterate LLM app workflows linking prompts and python code
- trace and debug interactions with LLMs
- evaluate flow quality and performance on datasets
- integrate LLM flow testing into CI/CD pipelines
- deploy prompt flows to a serving platform
- prototype chat-with-pdf style RAG applications

## When to choose
- you need structured orchestration and evaluation for LLM app development
- you want tracing and CI/CD quality gates for prompt-based flows
- you are already invested in the Azure AI ecosystem with existing flows

## When to avoid
- starting a new project - the project is retiring and Microsoft recommends migrating to Microsoft Agent Framework
- you need long-term supported tooling beyond April 2027
- you want a lightweight prompt library without workflow orchestration

## Facets
- artifact type: framework
- maturity: maintenance
- function: prompt-engineering, llm-inference, agent-framework, testing, workflow-automation, tracing, ci-cd
- domain: large-language-models, artificial-intelligence, developer-tools, machine-learning
- platform: python, cross-platform, cli, editor-plugin, cloud
- tags: llm-app-development, prompt-flow, evaluation, vscode-extension, azure-ai, orchestration, retiring, ai-agents

## Member repositories
- microsoft/promptflow (main) score 75

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:46.510322+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 2026-08-29T17:16:25.816642+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/promptflow (fetched 2026-08-28T04:10:46.510322+00:00, sha 9e324176dfd3)
  - homepage: https://microsoft.github.io/promptflow/ (fetched 2026-08-29T08:14:52.963286+00:00, sha dc34c3c3204a)
  - registry_pypi: https://pypi.org/pypi/promptflow/json (fetched 2026-08-29T08:14:53.058030+00:00, sha b12c2b94c229)
- Data as of 2026-08-30T08:39:29.467469+00:00.