Orchestrate review-first multi-agent development work with Kandev
Use Kandev when an operator needs to plan, run, isolate, review, and ship multiple coding-agent tasks through a controlled kanban and pipeline workspace.
npx skills add agentskillexchange/skills --skill orchestrate-review-first-multi-agent-development-work-with-kandev
Kandev is a self-hostable AI kanban and development environment for coordinating coding agents across providers. The operator defines tasks, assigns agents to pipeline steps, runs work in isolated worktrees or configured runtimes, inspects file diffs and conversations, and decides which changes become pull requests. It is useful when the work needs supervision, parallel execution, and traceable review rather than a single ad hoc agent session.
What this skill actually does
Invoke this skill when a team or solo operator wants a controlled agent-work queue: splitting feature work, running implementation and review agents side by side, tracking subtasks, managing remote executors, or keeping humans in the approval loop before shipping code. Use the product normally when someone just wants to browse its UI; use this skill when the agent is expected to operate the workflow and preserve review boundaries.
The scope boundary is orchestration of agent development tasks from intake through reviewable output. This is not a generic project-management product card or a broad coding-agent platform listing; the repeatable workflow is to create bounded tasks, choose agent/runtime execution, isolate changes, review artifacts, and open or withhold PRs.