From ork
Decomposes complex work into trackable tasks with dependency chains using TaskCreate, TaskUpdate, TaskGet, TaskList tools. For multi-step implementations, parallel coordination, and status tracking.
npx claudepluginhub yonatangross/orchestkit --plugin orkThis skill is limited to using the following tools:
Claude Code 2.1.16 introduces a native Task Management System with four tools:
checklists/dependency-checklist.mdchecklists/task-design-checklist.mdreferences/dependency-tracking.mdreferences/multi-agent-coordination.mdreferences/status-workflow.mdrules/_sections.mdrules/_template.mdrules/task-agent-coordination.mdrules/task-completion-criteria.mdrules/task-dependency-validation.mdscripts/task-tree-visualizer.pytest-cases.jsonDecomposes complex tasks into parallel units, designs dependency graphs with blockedBy/blocks, and coordinates multi-agent teams via task descriptions and workload balancing. For agent workflows and progress monitoring.
Manages tasks for session continuity, coordinating multi-step work, subagent assignments, and dependencies that survive compaction.
Orchestrates TASKS.md execution at scale with team (Agent Teams API, 3-level hierarchy), auto (task dispatch), or thin (minimal context) modes; auto-selects by task count and infrastructure.
Share bugs, ideas, or general feedback.
Claude Code 2.1.16 introduces a native Task Management System with four tools:
Tasks enable structured work tracking, parallel coordination, and clear progress visibility.
Break complex work into atomic, trackable units:
Feature: Add user authentication
Tasks:
#1. [pending] Create User model
#2. [pending] Add auth endpoints (blockedBy: #1)
#3. [pending] Implement JWT tokens (blockedBy: #2)
#4. [pending] Add auth middleware (blockedBy: #3)
#5. [pending] Write integration tests (blockedBy: #4)
Use addBlockedBy to create execution order:
// Task #3 cannot start until #1 and #2 complete
{"taskId": "3", "addBlockedBy": ["1", "2"]}
pending → in_progress → completed
↓ ↓
(unblocked) (active)
pending/in_progress → deleted
Use status: "deleted" to permanently remove tasks:
// Delete a task
{"taskId": "3", "status": "deleted"}
When to delete:
When NOT to delete:
Provide present-continuous form for spinner display:
| subject (imperative) | activeForm (continuous) |
|---|---|
| Run tests | Running tests |
| Update schema | Updating schema |
| Fix authentication | Fixing authentication |
Agent Teams provides multi-agent coordination with shared task lists and peer-to-peer messaging.
1. TeamCreate("my-feature") → Creates team + shared task list
2. TaskCreate(subject, description) → Add tasks to shared list
3. Agent(prompt, team_name, name) → Spawn teammates
4. TaskUpdate(owner: "teammate-name") → Assign tasks
5. SendMessage(type: "message") → Direct teammate communication
6. SendMessage(type: "shutdown_request") → Graceful shutdown
| Criteria | Task Tool (subagents) | Agent Teams |
|---|---|---|
| Independent tasks | Yes | Overkill |
| Cross-cutting changes | Limited | Yes |
| Agents need to talk | No (star topology) | Yes (mesh) |
| Cost sensitivity | Lower (~1x) | Higher (~2.5x) |
| Complexity < 3.0 | Yes | No |
| Complexity > 3.5 | Possible | Recommended |
# Spawn teammate into shared task list
Agent(
prompt="You are the backend architect...",
team_name="my-feature",
name="backend-architect",
subagent_type="backend-system-architect"
)
# Teammate claims and works tasks
TaskList → find unblocked, unowned tasks
TaskUpdate(taskId, owner: "backend-architect", status: "in_progress")
# ... do work ...
TaskUpdate(taskId, status: "completed")
TaskList → find next task
# Direct message between teammates
SendMessage(type: "message", recipient: "frontend-dev",
content: "API contract ready: GET /users/:id returns {...}",
summary: "API contract shared")
# Broadcast to all (use sparingly)
SendMessage(type: "broadcast",
content: "Breaking change: auth header format changed",
summary: "Breaking auth change")
When using Agent Teams, if context limit is reached mid-workflow:
TaskListTaskUpdate(status: "cancelled", note: "context limit")ork:implement - Implementation workflow with task tracking and progress updatesork:verify - Verification tasks and completion checklistsork:fix-issue - Issue resolution with hypothesis-based RCA trackingork:brainstorm - Design exploration with parallel agent tasksLoad on demand with Read("${CLAUDE_SKILL_DIR}/references/<file>"):
| File | Content |
|---|---|
dependency-tracking.md | Dependency tracking patterns |
status-workflow.md | Status workflow details |
multi-agent-coordination.md | Multi-agent coordination |