run-flow

Execute bounded multi-task workflows with risk-based verification and optional reviewer gate.

1|Updated Feb 3, 2026
One-click install
npx skills add https://github.com/bohewu/agents_pipeline --skill run-flow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: run-flow
Source: https://github.com/bohewu/agents_pipeline/tree/main/skills/run-flow
Command: npx skills add https://github.com/bohewu/agents_pipeline --skill run-flow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill streamlines the execution of complex multi-task workflows by ensuring bounded execution, risk-derived verification, and an optional reviewer gate, providing a controlled and safe way to manage multi-agent pipelines.

Core Features & Use Cases

  • Bounded Execution: Ensures that workflows execute a maximum of five tasks, reducing the risk of overruns and system strain.
  • Risk Verification: Before executing, verifies that the current workspace profile is healthy and eligible, ensuring safety and reliability.
  • Reviewer Gate: Optionally includes a reviewer gate for critical workflows, adding an extra layer of quality control.
  • Use Case: Ideal for engineering workflows where a series of tasks need to be executed in a controlled environment, such as software development or project management.

Quick Start

To run a flow with the 'run-flow' skill, use the command: run-flow

Frequently Asked Questions about run-flow

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute a multi-task workflow with bounded execution limits?

Bounded execution in a multi-task workflow limits pipeline runs to a maximum of five tasks, reducing system strain and preventing overruns. You can execute this controlled process using the run-flow command.

What is risk-based verification in a multi-agent pipeline?

Risk-based verification checks that the current workspace profile is healthy and eligible before executing tasks. This assessment ensures reliability and safety within AI-driven engineering and project management workflows.

Can I add a reviewer gate for critical project management workflows?

Yes, you can optionally include a reviewer gate for critical workflows. This adds an extra layer of quality control to the engineering pipeline before the bounded task execution proceeds.

What are the task limits when running a multi-agent pipeline?

The system enforces a strict limit of a maximum of five tasks per workflow execution. This bounded execution design reduces the risk of overruns and minimizes system strain during pipeline operations.

When should I use bounded execution for software engineering tasks?

Use bounded execution when managing complex engineering workflows where a series of tasks must run in a controlled environment. It ensures safety through risk-derived verification and prevents unbounded system overruns.