OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Workflow Agents

Manifold

Manifold is an experimental, MIT-licensed Go platform that coordinates teams of AI assistants across long-horizon workflows through specialist agents, visual workflow composition, projects, tools and scheduled tasks.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Manifold — OpenViST project profile in Workflow Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Manifold when a task needs extended, multi-step agent coordination rather than a one-shot prompt, especially when visual workflows, project isolation, MCP tools, scheduled execution or model-provider flexibility matter.

Who it is for

  • Developers building multi-step AI agent workflows
  • Teams coordinating specialist assistants across extended objectives
  • Operators connecting hosted or self-hosted language models
  • Builders needing visual workflow composition and MCP tool integration

Documented uses

  • Coordinate multi-step objectives with teams of AI assistants.
  • Assemble workflows visually and reuse saved workflows as tools or nodes.
  • Connect OpenAI, Google, Anthropic or OpenAI-compatible self-hosted models.
  • Create isolated project workspaces with project-specific skills.
  • Schedule specialist tasks and send results to supported external channels.
Before you choose

Requirements & limits

The documented first-run route uses Docker Compose and a local web interface. A basic deployment requires Docker, an LLM provider or reachable OpenAI-compatible endpoint, and a writable WORKDIR. Manifold is MIT-licensed and publicly available as source code.

  • The platform supports OpenAI, Google and Anthropic models plus OpenAI-compatible endpoints served through llama.cpp or vLLM.
  • Workflow design is visual, and MCP tools can be exposed as nodes.
  • Projects isolate agent workspaces and support project-specific skills.
  • Pulse supports interval, daily and one-time scheduled tasks.
  • The README labels Manifold experimental, which is relevant when evaluating deployment maturity.

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

Set up Manifold with Docker Compose

Prepare a basic local Manifold deployment and open its web interface.

Before you start

  • Docker with Docker Compose support
  • An LLM API key or reachable OpenAI-compatible endpoint
  • A writable absolute host directory for WORKDIR

Steps

  1. Copy example.env to .env and config.yaml.example to config.yaml.

  2. Edit .env and set at least OPENAI_API_KEY and WORKDIR.

  3. Run docker compose up -d manifold.

  4. Open http://localhost:32180.

Expected outcome

A local Manifold service is started through the documented Docker-based first-run path.

Guide limitations

  • This is a deployment setup path rather than a workflow execution test.
  • The project describes Manifold as experimental.
  • The documented path still requires an LLM provider, either through a remote API key or a reachable OpenAI-compatible endpoint.
Documentation 1 ↗
Evidence and transparency

Sources & review

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