The Long-running agent harness allows AI agents to efficiently work across multiple context windows with a structured approach. It's designed for developers and teams looking to manage incremental tasks and track progress across sessions effectively.
$ npx skills add https://github.com/michael-f-bryan/skills --skill long-running-agent-harnessThe Long-running agent harness is a skill designed for AI agents working across multiple context windows with structured task management. It enables developers to handle incremental tasks efficiently while tracking progress across multiple sessions. This harness provides a framework for managing extended workflows where context limitations require breaking work into manageable segments. Teams can install the skill using the skills CLI and optionally symlink it to their global agents folder for local development.
Install the skill using the command provided.
Plan and manage AI agent tasks incrementally.
Ensure each session has clear objectives and documentation.
Facilitate human check-ins at defined milestones.
$ npx skills add https://github.com/michael-f-bryan/skills --skill long-running-agent-harnessgit clone https://github.com/michael-f-bryan/skillsCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Check the GitHub repository or documentation for usage examples.
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan