synapse-ai

SynapseOrch: AI workflows that follow the plan. Describe a job, approve the steps, run them on any model, local or cloud. MCP server included. Open source (AGPL

What is synapse-ai?

SynapseOrch: AI workflows that follow the plan. Describe a job, approve the steps, run them on any model, local or cloud. MCP server included. Open source (AGPL-3.0).

Overview of synapse-ai

What it is

SynapseOrch is an open-source application that builds and runs AI workflows with a fixed path. Users describe a job in plain language, and the tool shows a plan including steps, models, cost estimates, and sign-off points. Nothing is built until the user approves the plan, and every run follows that approved path. Models perform tasks within steps but do not decide the sequence of steps. It is self-hosted, supports any model, and has no account requirement, license check, or telemetry. It includes a UI, REST API, CLI, and MCP server.

Key features

  • Synapse AI: an assistant that builds and modifies agents and workflows from a sentence.
  • Workflows: visual canvas with steps including agents, models, tools, code, branches, loops, parallel paths, sign-offs, and sub-workflows.
  • Agents and teams: agents with their own model, tools, skills, and memory; team agents split work among members.
  • Sign-offs: steps that wait for human approval, answered in the app or via chat channels (Slack, Telegram, etc.).
  • Knowledge: indexing of files and web pages with cited answers (optional).
  • Any model: support for 23+ providers, OpenAI-compatible endpoints, local models via Ollama, and CLI tools like Claude Code.
  • Cost tracking: estimate before a run, per-step cost after, and spending cap per run.

Use cases

Developers and automation builders who need reliable AI workflows that behave consistently across runs. Suitable for teams or clients requiring self-hosted solutions with no external dependencies. Use cases include scheduled tasks, multi-step data processing, agent collaboration with human oversight, and integration with chat platforms.

Getting started

Install via one-line command: bash curl -LsSf https://synapseorch.com/install.sh | bash # macOS, Linux powershell irm https://synapseorch.com/install.sh | iex # Windows (PowerShell) Alternatively, use npm, pip, or Docker: bash npm install -g synapse-orch-ai && synapse bash pip install synapse-orch-ai && synapse bash docker run -d --name synapse -p 3000:3000 -p 127.0.0.1:8765:8765 -v synapse-data:/data -v /var/run/docker.sock:/var/run/docker.sock synapseorchai/synapse-ai:latest After installation, open the provided link to start. The installer asks for a model key; a free Gemini key from Google AI Studio can be used.

License and pricing

Licensed under AGPL-3.0. Open source, no account or license check required.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/synapseorch-aisynapse-ai.svg)](https://yunodb.com/tools/synapseorch-aisynapse-ai?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/synapseorch-aisynapse-ai?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/synapseorch-aisynapse-ai.svg" alt="Featured on YunoDB" height="20"></a>