Verify tests, build, and linter before closing beads issues.
npx skills add https://github.com/kirvin/agent-dev-harness --skill task-completion
Please help me install this Agent Skill. Skill: task-completion Source: https://github.com/kirvin/agent-dev-harness/tree/main/plugins/kf/skills/task-completion Command: npx skills add https://github.com/kirvin/agent-dev-harness --skill task-completion
This Skill ensures thorough task completion by verifying requirements before closing beads issues and conducting a lessons-learned retro.
Run this skill to close beads issues with non-trivial changes, using the command 'close-task-completion <issue-id>'.
High-intent search queries and answers about installing and using this skill.
To verify tests and build status before closing a beads issue, you can use a task completion skill that checks tests, build, and linter status. This ensures all end-to-end feature functionality passes before final closure.
A lessons-learned retro is a post-mortem reflection conducted during issue closure to improve future workflows. You need it when closing beads issues with non-trivial changes to identify friction points and systemic issues.
Run a post-mortem retro by prompting for reflection on the completed task to document friction points and systemic issues. This process addresses workflow friction and generates improvement issues for future development cycles.
Yes, you can use automated verification for linter and build checks when completing a task. This verification ensures thorough closure by confirming tests, build, and linter pass before an issue is officially closed.
The best way to identify systemic issues during software issue closure is conducting a lessons-learned retro. It prompts post-mortem reflection to address friction points and create improvement issues for future workflows.
You should not use a post-mortem retro for task closure when dealing with trivial changes. This verification and retro process is designed specifically for closing beads issues that involve non-trivial changes and require thorough documentation.
jiayaoqijiaCoordinates four parallel agents to create source and test files in non-overlapping packages, then validates and commits them.
LerianStudioCapture portfolio-level lessons learned and produce retrospective artifacts.
efiadmTest mixin build scripts for correct integration and deterministic execution.
LGuguiCloses sessions via checklist, updates TAREFAS.md, KANBAN, and Paperclip.
