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

Spark

Spark is a self-hosted AI coding agent with a Python backend and browser interface. It works against a chosen project directory using file read and write, search, shell, and Jupyter tools, and supports multi-step plans, parallel sub-agents, persistent project memory, sandbox controls, and session checkpoints.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

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

Practical overview

When it fits

Use Spark when you want a coding agent to work against a project directory through a local browser interface, with configurable tool permissions and retained project context.

Who it is for

  • Developers who want an AI agent to investigate and edit a local project repository.
  • Teams or individuals who want to set tool permissions and approve agent actions.
  • Developers who need project-aware guidance retained across sessions.

Documented uses

  • Search project files and inspect code.
  • Ask the agent to make file changes or perform shell and Jupyter tasks.
  • Break a coding request into a multi-step plan and delegate subtasks to parallel sub-agents.
  • Apply project conventions through AGENTS.md and persistent memory.
  • Review earlier session states and roll back to a checkpoint.
Before you choose

Requirements & limits

The README documents a self-hosted setup, Python 3.11 or newer, and an OpenAI-compatible API or model backend. Node.js 18 or newer is required for frontend development only. The repository states that Spark is MIT-licensed.

  • Spark's primary interface is a browser served locally after running its web command.
  • Model connections use OpenAI-compatible APIs; named options include DeepSeek, OpenAI, Ollama, and Azure.
  • The four-level sandbox, approval flow, and protected paths are relevant when deciding how much access to grant.
  • MCP compatibility is stated, but the README does not document a particular server integration.

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 Spark's browser interface for a project

Install Spark and start its web interface pointed at a project directory.

Before you start

  • Python 3.11 or newer
  • An OpenAI-compatible API or model backend, such as DeepSeek, OpenAI, Ollama, or Azure
  • A project directory to use as the work directory

Steps

  1. Clone the Spark repository and enter its directory.

  2. Create and activate a Python virtual environment.

  3. Install Spark with the command below.

    Documented command
    pip install -e ".[dev]"

  4. Run spark web --workdir /your/project and open http://localhost:8000 in a browser.

Expected outcome

A local browser interface for conversing with Spark about the selected project.

Guide limitations

  • The README lists Node.js 18 or newer for frontend development only.
  • Provider-specific model setup is not included in the README.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for AI coding assistance

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
SparkSearch project files and inspect code.CLI, Self-hosted, Server, WebThe README documents a self-hosted setup, Python 3.11 or newer, and an OpenAI-compatible API or model backend. Node.js 18 or newer is required for frontend development only. The repository states that Spark is MIT-licensed.
LlmcodeInspect a project's directory structure with `/tree`.CLIThe documented setup requires an API key and supports an API base URL and model setting. The project is open source under Apache-2.0.
Klaat CodeOpen a project directory for terminal-based coding assistance.CLI, ServerThe README provides a global npm installation command and documents browser sign-in with a KlaatAI account. It also identifies the repository as open source under Apache 2.0.
Evidence and transparency

Sources & review

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