Multi-Agent Workflow Kit by laris-co is an orchestration framework designed to coordinate multiple specialized AI agents across automated software engineering and data workflows. It provides the scaffolding needed to define distinct agent roles, manage execution flow, and handle state transfer between different steps in a multi-agent pipeline. In practical application, the kit serves as a harness where developers configure individual sub-agents with discrete system prompts, tools, and model backends. The orchestrator routes tasks sequentially or conditionally based on step outcomes, coordinating outputs across terminal interfaces or backend script runners. Unlike monolithic CLI coding agents, Multi-Agent Workflow Kit emphasizes modularity and pipeline composition. Rather than relying on a single agent prompt to handle planning, implementation, and verification, it splits these concerns into configurable nodes that can be independently audited and modified. The project was originally hosted on GitHub as an open-source repository. Note that the upstream repository at github.com/laris-co/multi-agent-workflow-kit may currently be unavailable or relocated.
Multi-Agent Coordination — Orchestrates task handoffs and state sharing between distinct agent workers. - Modular Agent Roles — Configures specialized agents for distinct tasks like coding, testing, and validation. - Flexible Pipeline Routing — Supports deterministic and conditional task execution paths. - Tool Integration Layer — Provides hooks to equip agents with custom scripts and CLI execution tools. - Pluggable LLM Backends — Allows connecting different language models based on specific agent requirements.
Use cases
Use Case: Automated Code Generation and Review A developer configures a pipeline where an initial coding agent generates patch files from an issue description, and a secondary reviewer agent runs automated checks and validates diffs before finalizing changes.
Use Case: Multi-Step Refactoring Pipelines An engineering team sets up chained agents to handle large-scale code transformations, splitting the work into discovery,AST manipulation, and unit test execution agents.
Use Case: CI/CD Agentic Automation A DevOps engineer integrates the workflow kit into continuous integration jobs to run autonomous bug triage and regression report generation.
Who it is for
This harness fits software engineers and pipeline architects seeking to design multi-agent workflows with explicit task routing rather than single-agent chat assistants. It is less suitable for developers looking for an out-of-the-box interactive terminal coding assistant with a built-in TUI.
Install guide
Installation Clone the repository or install the package dependencies (standard Python environment setup): bash git clone https://github.com/laris-co/multi-agent-workflow-kit.git cd multi-agent-workflow-kit pip install -e .(Note: Check repository availability as the upstream URL may be relocated.)
First Run Configure your environment variables for your chosen LLM provider: bash export OPENAI_API_KEY="your-api-key" # or other provider credentials as required
Example Session Execute a multi-agent workflow definition via the runner: bash python -m workflow_kit.runner --config ./workflows/code_review.yaml
ActPlane — ActPlane is an open-source agent execution and policy enforcement harness developed by the Eunomia-bpf organization. It functions as a security…
hol-guard — hol-guard is a security scanner and agent guardrail harness developed by Hashgraph Online. Built to safeguard agentic runtime environments and…
Skillreaper — Skillreaper is an open-source skill lifecycle management and auditing tool for AI agent ecosystems, developed by thousandflowers. It analyzes agent…
Agent Island — Agent Island is an open-source agent orchestration harness created by developer tristan666666 on GitHub. Designed as a framework for running…
Stratless — Stratless is an open-source agent harness and execution orchestrator developed by the Stratless AI team. It provides a modular infrastructure…
Google AX — Google AX is an open-source agent harness and execution orchestrator developed by Google. It provides runtime infrastructure and execution sandboxing…