OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

Flowcraft

Flowcraft is a modular Go toolkit for building and evaluating AI applications and agents with execution, graph, tool, inference, memory, deployment and runtime components. It can be embedded as a library or used for daemon, real-time pipeline and local interactive workflows.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Flowcraft — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Flowcraft when an agent application needs Go-native modularity across execution, providers, memory and runtime, or when a local workspace is useful for exploring scenarios before embedding the core in a service.

Who it is for

  • Go developers building AI agents.
  • Teams integrating multiple model-provider wire families.
  • Engineers designing agent runtimes with memory, retrieval, sessions or checkpoints.
  • Developers evaluating local agent workspaces and scripted scenarios.

Documented uses

  • Assemble an AI agent runtime from modular Go components.
  • Add provider adapters without coupling application code to one model provider.
  • Connect long-term memory and knowledge retrieval through memory contracts and implementations.
  • Run agent turns through runtime sessions.
  • Create and test a local interactive Forge workspace.
Before you choose

Requirements & limits

The repository is public and identifies an MIT license. The project is presented as a self-hosted Go SDK and includes CLI and server-oriented runtime options.

  • Go 1.26 or later is indicated by the repository badge.
  • The core module provides contracts, while provider adapters and platform backends are separate modules.
  • Memory implementations are registered outside core and selected through implementation names and parameters.
  • The Forge example offers a documented local workspace path with scenario configuration, interactive use and scripted tests.
  • Applications can use direct core packages or declarative graphs; graphs are an available architecture rather than a requirement.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Create and test a Forge workspace

Set up the repository’s local Forge workspace from the werewolf scenario and run its documented scripted test.

Before you start

  • A Go 1.26 or later environment.
  • A local copy of the Flowcraft repository.
  • The repository’s examples/forge workspace and scenario files.

Steps

  1. Change into examples/forge.

  2. Run the help command to view the demo command reference.

  3. Create a workspace from the werewolf scenario with the documented workspace command.

  4. Run the documented werewolf/opening_setup scripted test.

Expected outcome

A Forge workspace is created from the werewolf scenario and the named scripted test is invoked.

Guide limitations

  • This recipe covers the documented local demo setup and test invocation, not a production deployment.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
3 October 2026
Original website title
GizClaw/flowcraft · Official homepage title · retained in its source language
How we review sourcesSuggest a correction