Automate iterative coding task execution and verification with feedback loops.
npx skills add https://github.com/gaasher/Agent-Loop-Skills --skill swe-loop
Please help me install this Agent Skill. Skill: swe-loop Source: https://github.com/gaasher/Agent-Loop-Skills/tree/main/loops/swe-loop Command: npx skills add https://github.com/gaasher/Agent-Loop-Skills --skill swe-loop
This Skill automates the execution of a coding loop, handling task management, execution, and quality control. It streamlines the development process by decomposing a coding prompt into an executable plan and managing the execution and verification of each task.
To implement a coding prompt using swe-loop, use the command 'swe-loop run <prompt> <repo>'
High-intent search queries and answers about installing and using this skill.
Automating the coding loop for software repository tasks involves decomposing a prompt into an executable plan, executing tasks iteratively, and verifying them with feedback loops. This approach manages writing, testing, and maintaining code quality with human oversight.
Rubric-based quality control in iterative code execution evaluates tasks using both objective metrics like passing tests and qualitative rubric assessments. It ensures each coding loop iteration meets defined quality standards before proceeding to the next development phase.
Yes, automated repository task management requires Python 3.9 or higher. You also need a valid test command configured for your repository to enable the automated execution and verification of coding tasks.
Yes, automated coding loops are applicable to tasks requiring multiple rounds of development and quality assurance, including both code fixes and feature implementations. The loop handles prompt refinement, execution, and verification within a real repository.
To start automating a coding prompt, use the command 'swe-loop run <prompt> <repo>'. This initiates the automated coding loop, handling task management, execution, and quality control for your software repository.
Automated loops require a valid test command and Python 3.9+ to function, limiting use in repositories without test setups. Additionally, while it automates execution and quality control, human oversight is still needed for complex or unclear cases.
final-runAutomate repo-local FinalRun CLI install, validation, and test execution.
LuminairPrimePush code changes and create pull requests with validation checks.
791994545Decompose complex tasks into structured plans with loop commands.
tech-with-sethCreate Vitest unit tests co-located with source files using project templates.
