toolsSource-backedReview first Safety ✓ Privacy ✓

Prompt flow

Open-source suite of development tools from Microsoft for building LLM applications end to end — create executable flows that link LLMs, prompts, Python, and tools, trace and debug them, evaluate quality against datasets in CI/CD, and deploy to a serving platform.

by microsoft · submitted by davion-knight·added 2026-07-10·

CLI

HarnessCLI

Command center

Source

Review first

Review safety and privacy notes before installing or copying commands.

Safety notes Privacy notes

Install & copy

## Editorial notes

Prompt flow is useful when Claude-adjacent teams want to treat LLM features like real software — building them as testable flows, tracing what the model did, and measuring quality before shipping. It is a suite of development tools from Microsoft that streamlines the end-to-end lifecycle of LLM applications, from prototyping through evaluation to deployment and monitoring.

This is distinct from the agent frameworks, memory, and search tools in the directory: rather than a runtime agent library, Prompt flow is the development, evaluation, and deployment toolkit for LLM app flows, with a VS Code experience and CI/CD-friendly evaluation.

## Key capabilities

- **Executable flows** — create flows that link LLMs, prompts, Python code, and other tools into a runnable graph.
- **Tracing and debugging** — trace interactions with LLMs to debug and iterate on flows.
- **Evaluation** — evaluate flow quality and performance against larger datasets, with metrics you can extend.
- **CI/CD integration** — fold testing and evaluation into a CI/CD pipeline to guard flow quality over time.
- **Deployment** — deploy a flow to a serving platform or integrate it into an application's codebase.
- **Connections** — manage model-provider and tool credentials as connections used by flows.
- **VS Code experience** — a VS Code extension for authoring, running, and visualizing flows.
- **Local or cloud** — run locally, with an optional Azure AI cloud version for team collaboration.

## How teams use it

- **Prototype to production** — take an LLM feature from a first prototype to a tested, deployable flow.
- **Quality gates** — run evaluations in CI/CD so prompt or model changes cannot silently regress quality.
- **Debugging** — trace a flow's LLM calls to understand and fix unexpected behavior.
- **Prompt engineering** — iterate on prompts and flow structure with a visual, testable workflow.
- **Deployment** — serve a flow as an endpoint or embed it in an application.

## Getting started

Prompt flow is open source and runs locally. Install the SDK and CLI with `pip install promptflow`,
optionally add the VS Code extension, and configure connections for your model providers. Author a
flow that links prompts, LLMs, and Python nodes, trace and debug it, evaluate it against a dataset,
and then deploy it to a serving target or integrate it into your app; an Azure AI cloud version is
available for team collaboration.

## Source notes

- The official repository describes Prompt flow as a suite of development tools for the end-to-end development cycle of LLM-based AI applications, from ideation and prototyping through testing, evaluation, deployment, and monitoring.
- Documented capabilities include creating executable flows that link LLMs, prompts, Python code, and tools; debugging and iterating with tracing of LLM interactions; evaluating flow quality and performance against larger datasets; integrating testing and evaluation into CI/CD; and deploying flows to a serving platform or an application codebase.
- Prompt flow provides a Python SDK and CLI and a VS Code extension, and offers an optional cloud version through Azure AI for team collaboration.
- The GitHub repository is `microsoft/promptflow`, is MIT licensed, is installed from PyPI as `promptflow`, and is maintained by Microsoft.

## Duplicate check

Checked current `content/tools/`, `content/mcp/`, agents, skills, hooks, rules, commands, guides, open pull requests, and repository-wide content for `Prompt flow`, `promptflow`, `microsoft/promptflow`, `microsoft.github.io/promptflow`, `LLM app development`, and `LLM evaluation`. Existing entries cover adjacent agent frameworks and evaluation tools, but no dedicated Prompt flow tools entry, Prompt flow source URL duplicate, or open duplicate PR was found.

## Disclosure

Editorial listing. No paid placement or affiliate link is used.

Trust & readiness

TrustReview first
Sourcesource-backed
Safety notesPresent
ReviewedYes

Community context

Related entries(4)
Related guides(3)
Community signals

Compare

Integrations & API

Contribute

Suggest a metadata change Claim this listing

Documentation Source repository Browse directory

Review first — review before installing

Open the source and read safety notes before installing.

Citation facts

Source-backed facts for citing this resource, derived directly from the registry — also available as plain text for AI assistants.

Canonical URL: https://heyclau.de/entry/tools/prompt-flow
Source URLs: https://microsoft.github.io/promptflow/, https://github.com/microsoft/promptflow
Brand: Prompt flow
Brand domain: microsoft.github.io
Brand asset source: brandfetch
Safety notes: Flows execute nodes that run LLM prompts, Python code, and tools, so review what a flow does before running it, especially flows from untrusted sources or over untrusted input., Connections store model-provider credentials; scope them to the minimum needed and keep connection configuration out of source control., Treat flow inputs and LLM outputs as untrusted for downstream actions, and gate any node that writes data or calls external services., Evaluation and tracing capture prompts, inputs, and outputs; confirm what is recorded and where before running on sensitive data., Keep production flows, connections, and permissions narrower than sample flows and notebooks.
Privacy notes: Running a flow sends prompts and inputs to the configured model providers, which process that data under their own terms., Traces, batch runs, and evaluation results can record prompts, inputs, outputs, and metadata, so choose retention and access controls for where those are stored., Connection secrets, evaluation datasets, and exported run data should be treated as sensitive and kept out of version control., The optional Azure AI cloud version processes flow and run data under its terms; running locally keeps that data in your environment.
Author: microsoft
Submitted by: davion-knight
Claim status: unclaimed
Last verified: 2026-07-10

Decision playbook

Review trust signals before you adopt

Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.

Compare context

Selected

Current score

Baseline

—

Delta

No baseline selected

No major trust-signal divergence detected in the current selection.

Source and provenance checks

Complete

Confirm ownership and provenance before trusting install instructions.

Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
Registry metadata indicates a reviewed listing.
Done

Safety and privacy checks

Complete

Validate risk disclosures before installation or API wiring.

Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done

Package and install checks

Needs review

Check package metadata and artifact integrity signals.

Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending

Compare-driven decision checks

Needs review

Use compare context to validate trade-offs before adoption.

Compare tray has multiple entries
Add at least one more entry to compare trust differences.
Pending
Baseline comparison available
No baseline peer selected yet.
Pending
Diverging trust signals identified
No major trust-signal divergence found.
Pending

Adoption plan

Balanced adoption plan

Current risk score 16/100. Use staged verification before broader rollout.

Risk 16

Pre-adoption checks

Validate source and review signals before any execution.

Confirm source provenanceRequired
Source URL/provenance metadata is present.
Done
Confirm metadata review state
Listing has review metadata.
Done
Verify install payload
Install/config payload exists and can be inspected.
Done

Security checks

Confirm safety, privacy, and package integrity signals.

Review safety notesRequired
Safety notes are present.
Done
Review privacy notesRequired
Privacy notes are present.
Done
Verify package integrity metadata
No package verification/checksum metadata.
Pending

Rollout

Adopt in controlled steps based on the selected plan.

Run in isolated sandbox firstRequired
Use a constrained sandbox and observe behavior across multiple tasks.
Pending
Roll out graduallyRequired
Roll out to a small cohort before wider usage.
Pending
Set monitoring and fallback
Define rollback path and monitor errors after adoption.
Pending

Evidence readiness

Evidence readiness matrix · balanced

Required evidence gates are covered (5/6 signals complete).

Risk 15

Source provenance

Present

Source repository/provenance is listed.

Required in this preset

Metadata review

Present

Review metadata is present.

Required in this preset

Safety notes

Present

Safety notes are present.

Required in this preset

Privacy notes

Present

Privacy notes are present.

Optional in this preset

Package integrity

Missing

Package integrity metadata is missing.

Optional in this preset

Install payload

Present

Install payload is available.

Required in this preset

Required evidence gates are covered for this preset.

Decision timeline

Decision timeline · balanced

5/6 steps complete with no blocking gaps for this preset.

Risk 14

triage

Confirm source provenanceRequired

Source/provenance metadata is available.

Done

triage

Check metadata review statusRequired

Review metadata is available.

Done

verify

Review safety notesRequired

Safety notes are available.

Done

verify

Review privacy notes

Privacy notes are available.

Done

verify

Validate package integrity metadata

Package integrity metadata is missing.

Pending

rollout

Verify install payload and commandsRequired

Install payload is available.

Done

No required blockers for this timeline preset.

Prerequisite readiness

5 prerequisites to line up before setup. Have accounts and credentials ready first.

0/5 ready

Account & credentials2Install & runtime1General2

Safety notes

Flows execute nodes that run LLM prompts, Python code, and tools, so review what a flow does before running it, especially flows from untrusted sources or over untrusted input.
Connections store model-provider credentials; scope them to the minimum needed and keep connection configuration out of source control.
Treat flow inputs and LLM outputs as untrusted for downstream actions, and gate any node that writes data or calls external services.
Evaluation and tracing capture prompts, inputs, and outputs; confirm what is recorded and where before running on sensitive data.
Keep production flows, connections, and permissions narrower than sample flows and notebooks.

Privacy notes

Running a flow sends prompts and inputs to the configured model providers, which process that data under their own terms.
Traces, batch runs, and evaluation results can record prompts, inputs, outputs, and metadata, so choose retention and access controls for where those are stored.
Connection secrets, evaluation datasets, and exported run data should be treated as sensitive and kept out of version control.
The optional Azure AI cloud version processes flow and run data under its terms; running locally keeps that data in your environment.

Prerequisites

Python project and a package manager to install the `promptflow` SDK and CLI from PyPI (a VS Code extension is also available).
Model-provider credentials configured as connections for the LLMs your flows call.
Test and evaluation datasets with expected outputs if you want to measure flow quality.
A serving target or application codebase for deployment, and optionally an Azure AI workspace for the cloud version.
A plan for who can run flows, view traces, and access connections and evaluation data.

Schema details

Install type: copy
Troubleshooting: No

Source repository stats

Scope: Source repo

Tool listing metadata

Website: https://microsoft.github.io/promptflow/
Pricing: open-source
Disclosure: editorial
Application category: DeveloperApplication
Operating system: macOS, Windows, Linux

Full copyable content

## Editorial notes

Prompt flow is useful when Claude-adjacent teams want to treat LLM features like real software — building them as testable flows, tracing what the model did, and measuring quality before shipping. It is a suite of development tools from Microsoft that streamlines the end-to-end lifecycle of LLM applications, from prototyping through evaluation to deployment and monitoring.

This is distinct from the agent frameworks, memory, and search tools in the directory: rather than a runtime agent library, Prompt flow is the development, evaluation, and deployment toolkit for LLM app flows, with a VS Code experience and CI/CD-friendly evaluation.

## Key capabilities

- **Executable flows** — create flows that link LLMs, prompts, Python code, and other tools into a runnable graph.
- **Tracing and debugging** — trace interactions with LLMs to debug and iterate on flows.
- **Evaluation** — evaluate flow quality and performance against larger datasets, with metrics you can extend.
- **CI/CD integration** — fold testing and evaluation into a CI/CD pipeline to guard flow quality over time.
- **Deployment** — deploy a flow to a serving platform or integrate it into an application's codebase.
- **Connections** — manage model-provider and tool credentials as connections used by flows.
- **VS Code experience** — a VS Code extension for authoring, running, and visualizing flows.
- **Local or cloud** — run locally, with an optional Azure AI cloud version for team collaboration.

## How teams use it

- **Prototype to production** — take an LLM feature from a first prototype to a tested, deployable flow.
- **Quality gates** — run evaluations in CI/CD so prompt or model changes cannot silently regress quality.
- **Debugging** — trace a flow's LLM calls to understand and fix unexpected behavior.
- **Prompt engineering** — iterate on prompts and flow structure with a visual, testable workflow.
- **Deployment** — serve a flow as an endpoint or embed it in an application.

## Getting started

Prompt flow is open source and runs locally. Install the SDK and CLI with `pip install promptflow`,
optionally add the VS Code extension, and configure connections for your model providers. Author a
flow that links prompts, LLMs, and Python nodes, trace and debug it, evaluate it against a dataset,
and then deploy it to a serving target or integrate it into your app; an Azure AI cloud version is
available for team collaboration.

## Source notes

- The official repository describes Prompt flow as a suite of development tools for the end-to-end development cycle of LLM-based AI applications, from ideation and prototyping through testing, evaluation, deployment, and monitoring.
- Documented capabilities include creating executable flows that link LLMs, prompts, Python code, and tools; debugging and iterating with tracing of LLM interactions; evaluating flow quality and performance against larger datasets; integrating testing and evaluation into CI/CD; and deploying flows to a serving platform or an application codebase.
- Prompt flow provides a Python SDK and CLI and a VS Code extension, and offers an optional cloud version through Azure AI for team collaboration.
- The GitHub repository is `microsoft/promptflow`, is MIT licensed, is installed from PyPI as `promptflow`, and is maintained by Microsoft.

## Duplicate check

Checked current `content/tools/`, `content/mcp/`, agents, skills, hooks, rules, commands, guides, open pull requests, and repository-wide content for `Prompt flow`, `promptflow`, `microsoft/promptflow`, `microsoft.github.io/promptflow`, `LLM app development`, and `LLM evaluation`. Existing entries cover adjacent agent frameworks and evaluation tools, but no dedicated Prompt flow tools entry, Prompt flow source URL duplicate, or open duplicate PR was found.

## Disclosure

Editorial listing. No paid placement or affiliate link is used.

About this resource

Editorial notes

Prompt flow is useful when Claude-adjacent teams want to treat LLM features like real software — building them as testable flows, tracing what the model did, and measuring quality before shipping. It is a suite of development tools from Microsoft that streamlines the end-to-end lifecycle of LLM applications, from prototyping through evaluation to deployment and monitoring.

This is distinct from the agent frameworks, memory, and search tools in the directory: rather than a runtime agent library, Prompt flow is the development, evaluation, and deployment toolkit for LLM app flows, with a VS Code experience and CI/CD-friendly evaluation.

Key capabilities

Executable flows — create flows that link LLMs, prompts, Python code, and other tools into a runnable graph.
Tracing and debugging — trace interactions with LLMs to debug and iterate on flows.
Evaluation — evaluate flow quality and performance against larger datasets, with metrics you can extend.
CI/CD integration — fold testing and evaluation into a CI/CD pipeline to guard flow quality over time.
Deployment — deploy a flow to a serving platform or integrate it into an application's codebase.
Connections — manage model-provider and tool credentials as connections used by flows.
VS Code experience — a VS Code extension for authoring, running, and visualizing flows.
Local or cloud — run locally, with an optional Azure AI cloud version for team collaboration.

How teams use it

Prototype to production — take an LLM feature from a first prototype to a tested, deployable flow.
Quality gates — run evaluations in CI/CD so prompt or model changes cannot silently regress quality.
Debugging — trace a flow's LLM calls to understand and fix unexpected behavior.
Prompt engineering — iterate on prompts and flow structure with a visual, testable workflow.
Deployment — serve a flow as an endpoint or embed it in an application.

Getting started

Prompt flow is open source and runs locally. Install the SDK and CLI with pip install promptflow, optionally add the VS Code extension, and configure connections for your model providers. Author a flow that links prompts, LLMs, and Python nodes, trace and debug it, evaluate it against a dataset, and then deploy it to a serving target or integrate it into your app; an Azure AI cloud version is available for team collaboration.

Source notes

The official repository describes Prompt flow as a suite of development tools for the end-to-end development cycle of LLM-based AI applications, from ideation and prototyping through testing, evaluation, deployment, and monitoring.
Documented capabilities include creating executable flows that link LLMs, prompts, Python code, and tools; debugging and iterating with tracing of LLM interactions; evaluating flow quality and performance against larger datasets; integrating testing and evaluation into CI/CD; and deploying flows to a serving platform or an application codebase.
Prompt flow provides a Python SDK and CLI and a VS Code extension, and offers an optional cloud version through Azure AI for team collaboration.
The GitHub repository is microsoft/promptflow, is MIT licensed, is installed from PyPI as promptflow, and is maintained by Microsoft.

Duplicate check

Checked current content/tools/, content/mcp/, agents, skills, hooks, rules, commands, guides, open pull requests, and repository-wide content for Prompt flow, promptflow, microsoft/promptflow, microsoft.github.io/promptflow, LLM app development, and LLM evaluation. Existing entries cover adjacent agent frameworks and evaluation tools, but no dedicated Prompt flow tools entry, Prompt flow source URL duplicate, or open duplicate PR was found.

Disclosure

Editorial listing. No paid placement or affiliate link is used.

#llm-apps #evaluation #tracing #prompt-engineering #python #workflows

Source citations

Source methodology →

Add this badge to your README

Show that Prompt flow is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.

[![Listed on HeyClaude](https://heyclau.de/badge/tools/prompt-flow.svg)](https://heyclau.de/entry/tools/prompt-flow)

How it compares

Prompt flow side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.

1 trust signal differ across this comparison (Submitter).

Field	Prompt flow Open-source suite of development tools from Microsoft for building LLM applications end to end — create executable flows that link LLMs, prompts, Python, and tools, trace and debug them, evaluate quality against datasets in CI/CD, and deploy to a serving platform. Open dossier	LangGraph Agent orchestration framework for building stateful, controllable, multi-step LLM and agent workflows. Open dossier	Microsoft AutoGen Open-source framework for building multi-agent AI applications, conversations, workflows, and autonomous systems. Open dossier	MLflow Open-source AI engineering platform for tracing, evaluating, prompt-managing, and deploying agents, LLM applications, and ML models. Open dossier
Next steps	Open dossier API JSON Open LLM Open source Newsletter Claim listing	Open dossier API JSON Open LLM Open source Newsletter Claim listing	Open dossier API JSON Open LLM Open source Newsletter Claim listing	Open dossier API JSON Open LLM Open source Newsletter Claim listing
Trust
Review status	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed
Package trust	Package not verified	Package not verified	Package not verified	Package not verified
Source provenance	Source-backed	Source-backed	Source-backed	Source-backed
SubmitterDiffers	davion-knight	—	—	oktofeesh1
Install risk	Review first	Review first	Review first	Review first
Notes	Safety ✓ Privacy ✓	Safety · Privacy ✓	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓
Brand	Prompt flow	LangGraph	Microsoft	MLflow
Category	tools	tools	tools	tools
Source	source-backed	source-backed	source-backed	source-backed
Author	microsoft	LangChain	Microsoft	MLflow Project
Added	2026-07-10	2026-04-27	2026-04-27	2026-06-03
Platforms	CLI	CLI	CLI	CLI
Source repo	—	—	—	—
Safety notes	✓Flows execute nodes that run LLM prompts, Python code, and tools, so review what a flow does before running it, especially flows from untrusted sources or over untrusted input. Connections store model-provider credentials; scope them to the minimum needed and keep connection configuration out of source control. Treat flow inputs and LLM outputs as untrusted for downstream actions, and gate any node that writes data or calls external services. Evaluation and tracing capture prompts, inputs, and outputs; confirm what is recorded and where before running on sensitive data. Keep production flows, connections, and permissions narrower than sample flows and notebooks.	— missing	✓AutoGen runs multi-agent workflows that can execute code and call external tools autonomously; sandbox execution and review agent actions before granting tool or system access.	✓MLflow evaluations, traces, judges, and dashboards are review signals, not proof that an agent, LLM application, prompt, model, or deployment is correct, safe, fair, or production-ready. Autologging, decorators, OpenTelemetry ingestion, manual spans, and framework integrations can wrap live application code and record intermediate agent steps, retrievals, tool calls, model requests, and model responses. LLM-as-a-judge scorers and prompt optimization workflows can call configured model providers, consume quota, hit rate limits, and produce evaluator-model errors that require separate handling. AI Gateway and serving workflows can centralize model access, routing, rate limits, and credentials; incorrect configuration can route traffic to the wrong provider or expose more access than intended. Production tracing, async logging, tracking servers, registries, artifact stores, and deployment endpoints should be reviewed for authentication, TLS, network exposure, backups, and incident response before production use. Model registry and deployment workflows can influence real production behavior, so promotion, rollback, and approval rules should be separated from exploratory eval results.
Privacy notes	✓Running a flow sends prompts and inputs to the configured model providers, which process that data under their own terms. Traces, batch runs, and evaluation results can record prompts, inputs, outputs, and metadata, so choose retention and access controls for where those are stored. Connection secrets, evaluation datasets, and exported run data should be treated as sensitive and kept out of version control. The optional Azure AI cloud version processes flow and run data under its terms; running locally keeps that data in your environment.	✓LangGraph sends prompts and graph state to your configured model provider (including Claude); persisted state and checkpoints can contain message and tool-call data.	✓AutoGen agents send prompts, code, and tool outputs to the configured LLM provider(s); review what data your agents transmit and each provider's data-handling and retention terms.	✓MLflow traces and evaluations can capture prompts, completions, retrieved context, tool arguments, tool outputs, spans, metadata, latency, token usage, costs, scores, datasets, expectations, and human feedback. Agent traces may contain customer data, private documents, source snippets, proprietary prompts, internal identifiers, secrets accidentally passed to tools, or model outputs that need redaction before storage or sharing. LLM-as-a-judge scorers, prompt optimization, AI Gateway calls, and serving endpoints may send prompts, outputs, context, or traces to configured model providers unless a reviewed local or private provider path is used. Tracking servers, backend databases, artifact stores, evaluation datasets, prompt registries, model registries, and exported reports should follow normal access-control, retention, audit-log, and deletion policies. Public demos, notebooks, and examples should not be copied into production workflows with real API keys, raw customer traces, unreleased prompts, or sensitive evaluation data.
Prerequisites	Python project and a package manager to install the `promptflow` SDK and CLI from PyPI (a VS Code extension is also available). Model-provider credentials configured as connections for the LLMs your flows call. Test and evaluation datasets with expected outputs if you want to measure flow quality. A serving target or application codebase for deployment, and optionally an Azure AI workspace for the cloud version.	— none listed	— none listed	Python environment, package manager, or managed MLflow environment for installing and running MLflow in the project being traced or evaluated. AI agent, LLM application, RAG pipeline, prompt workflow, model pipeline, or production trace source to connect to MLflow. MLflow tracking server, backend store, artifact store, or managed service path sized for traces, datasets, prompts, model artifacts, and evaluation results. Model provider credentials, gateway policy, rate limits, and budget controls for LLM calls, LLM-as-a-judge scorers, prompt optimization, and deployed endpoints.
Install	—	—	—	—
Config	—	—	—	—
Citations	Source repositorygithub.com 2026-07-10T08:39:04+00:00 Documentationmicrosoft.github.io Submitted by davion-knight2026-07-10 Source methodology →	Source repositorygithub.com 2026-07-10T08:39:04+00:00 Documentationlangchain-ai.github.io Source methodology →	Source repositorygithub.com 2026-07-10T08:39:04+00:00 Documentationmicrosoft.github.io Source methodology →	Source repositorygithub.com 2026-07-10T08:39:04+00:00 Documentationmlflow.org Submitted by oktofeesh12026-06-03 Source methodology →
Claim	Unclaimed	Unclaimed	Unclaimed	Unclaimed