Agent skill
swarm-operations
API reference for Claude Code swarm tools -- TeamCreate, SendMessage, TeamDelete, and Agent tool parameters. Use when looking up tool signatures, message schemas, shutdown sequences, error handling, or debugging swarm operations.
Install this agent skill to your Project
npx add-skill https://github.com/Jamie-BitFlight/claude_skills/tree/main/.claude/skills/swarm-operations
SKILL.md
Swarm Operations
API reference for multi-agent orchestration tools in Claude Code v2.1.45.
Tools Overview
flowchart TD
Start([What do you need to do?]) --> Q1{Operation type?}
Q1 -->|Create a team| TC[TeamCreate]
Q1 -->|Send a message| SM[SendMessage]
Q1 -->|Clean up team| TD[TeamDelete]
Q1 -->|Spawn an agent| Task[Agent tool]
Q1 -->|Manage work items| TT[TaskCreate / TaskUpdate / TaskList / TaskGet]
TeamCreate -- Create a Team
TeamCreate({
team_name: "feature-auth",
description: "Implementing OAuth2 authentication"
})
Creates:
~/.claude/teams/feature-auth/config.json~/.claude/tasks/feature-auth/directory- You become the team leader
SendMessage -- All Inter-Agent Communication
SendMessage handles direct messages, broadcasts, shutdown, and plan approval.
Direct Message (type "message")
SendMessage({
type: "message",
recipient: "security-reviewer",
content: "Please prioritize the authentication module. The deadline is tomorrow.",
summary: "Prioritize auth module"
})
Required fields: type, recipient, content, summary
IMPORTANT: Your text output is NOT visible to teammates. You MUST use SendMessage to communicate.
Broadcast (type "broadcast")
SendMessage({
type: "broadcast",
content: "Status check: Please report your progress",
summary: "Requesting status from all"
})
Required fields: type, content, summary
WARNING: Broadcasting is expensive -- sends N separate messages for N teammates. Prefer direct message to specific teammates.
When to broadcast:
- Critical issues requiring immediate team-wide attention
- Major announcements affecting everyone
When NOT to broadcast:
- Responding to one teammate
- Normal back-and-forth
- Information relevant to only some teammates
Shutdown Request (type "shutdown_request")
Leader requests teammate to exit:
SendMessage({
type: "shutdown_request",
recipient: "security-reviewer",
content: "All tasks complete, wrapping up"
})
Shutdown Response (type "shutdown_response")
Teammate approves or rejects shutdown:
// Approve (terminates your process)
SendMessage({
type: "shutdown_response",
request_id: "shutdown-123",
approve: true
})
// Reject (continue working)
SendMessage({
type: "shutdown_response",
request_id: "shutdown-123",
approve: false,
content: "Still working on task #3, need 5 more minutes"
})
IMPORTANT: Extract the requestId from the JSON shutdown_request message and pass it as request_id. Simply saying "I'll shut down" is not enough -- you must call the tool.
Plan Approval (type "plan_approval_response")
Leader approves or rejects teammate's plan:
// Approve
SendMessage({
type: "plan_approval_response",
request_id: "plan-456",
recipient: "architect",
approve: true
})
// Reject with feedback
SendMessage({
type: "plan_approval_response",
request_id: "plan-456",
recipient: "architect",
approve: false,
content: "Please add error handling for the API calls and consider rate limiting"
})
TeamDelete -- Remove Team Resources
TeamDelete()
Removes:
~/.claude/teams/{team-name}/directory~/.claude/tasks/{team-name}/directory
IMPORTANT: Will fail if teammates are still active. Use shutdown_request first.
Task Tool -- Spawn Teammates
Agent({
team_name: "my-project", // Required -- which team to join
name: "worker-1", // Required -- teammate's name
subagent_type: "general-purpose",
prompt: "Your instructions here...",
run_in_background: true // Teammates usually run in background
})
For spawning without team membership (subagents), see Skill(skill: "swarm-spawning").
Message Formats
Messages are JSON objects stored in inbox files.
Regular Message
{
"from": "team-lead",
"text": "Please prioritize the auth module",
"timestamp": "2026-01-25T23:38:32.588Z",
"read": false
}
Shutdown Request
{
"type": "shutdown_request",
"requestId": "shutdown-abc123@worker-1",
"from": "team-lead",
"reason": "All tasks complete",
"timestamp": "2026-01-25T23:38:32.588Z"
}
Shutdown Approved
{
"type": "shutdown_approved",
"requestId": "shutdown-abc123@worker-1",
"from": "worker-1",
"paneId": "%5",
"backendType": "in-process",
"timestamp": "2026-01-25T23:39:00.000Z"
}
Idle Notification (auto-sent when teammate stops)
{
"type": "idle_notification",
"from": "worker-1",
"timestamp": "2026-01-25T23:40:00.000Z",
"completedTaskId": "2",
"completedStatus": "completed"
}
Task Completed
{
"type": "task_completed",
"from": "worker-1",
"taskId": "2",
"taskSubject": "Review authentication module",
"timestamp": "2026-01-25T23:40:00.000Z"
}
Plan Approval Request
{
"type": "plan_approval_request",
"from": "architect",
"requestId": "plan-xyz789",
"planContent": "# Implementation Plan\n\n1. ...",
"timestamp": "2026-01-25T23:41:00.000Z"
}
Permission Request (for sandbox/tool permissions)
{
"type": "permission_request",
"requestId": "perm-123",
"workerId": "worker-1@my-project",
"workerName": "worker-1",
"workerColor": "#4A90D9",
"toolName": "Bash",
"toolUseId": "toolu_abc123",
"description": "Run npm install",
"input": {"command": "npm install"},
"permissionSuggestions": ["Bash(npm *)"],
"createdAt": 1706000000000
}
Error Handling
Common Errors
- "Cannot cleanup with active members" -- Teammates still running. Send shutdown_request to all teammates first, wait for approval.
- "Already leading a team" -- Team already exists. Call TeamDelete first, or use different team name.
- "Agent not found" -- Wrong teammate name. Read
config.jsonfor actual names. - "Team does not exist" -- No team created. Call TeamCreate first.
- "team_name is required" -- Missing team context. Provide
team_nameparameter. - "Agent type not found" -- Invalid subagent_type. Check available agents with proper prefix.
Graceful Shutdown Sequence
Always follow this sequence:
flowchart TD
A[1. Send shutdown_request to all teammates] --> B[2. Wait for shutdown_response approve]
B --> C[3. Verify no active members in config.json]
C --> D[4. Call TeamDelete]
// 1. Request shutdown for all teammates
SendMessage({ type: "shutdown_request", recipient: "worker-1", content: "All tasks complete" })
SendMessage({ type: "shutdown_request", recipient: "worker-2", content: "All tasks complete" })
// 2. Wait for shutdown approvals
// Messages arrive automatically in your inbox
// 3. Verify no active members
// Read ~/.claude/teams/{team}/config.json
// 4. Only then cleanup
TeamDelete()
Handling Crashed Teammates
Teammates have a 5-minute heartbeat timeout. If a teammate crashes:
- They are automatically marked as inactive after timeout
- Their tasks remain in the task list
- Another teammate can claim their tasks
- TeamDelete will work after timeout expires
Debugging
Read team config to see members:
# Check team config -- view members and backend types
# Read ~/.claude/teams/{team}/config.json
# Check teammate inboxes
# Read ~/.claude/teams/{team}/inboxes/{agent}.json
# List all teams
# ls ~/.claude/teams/
# Check task states
# Read ~/.claude/tasks/{team}/*.json
Related Skills
- Core concepts --
Skill(skill: "swarm-primitives") - Spawning agents --
Skill(skill: "swarm-spawning") - Patterns and recipes --
Skill(skill: "swarm-patterns")
SOURCE: Claude Code v2.1.45 tool descriptions (TeamCreate, SendMessage, TeamDelete) -- verified from system prompt 2026-02-18
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
ccc
This skill should be used when code search is needed (whether explicitly requested or as part of completing a task), when indexing the codebase after changes, or when the user asks about ccc, cocoindex-code, or the codebase index. Trigger phrases include 'search the codebase', 'find code related to', 'update the index', 'ccc', 'cocoindex-code'.
agent-browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.
delegate
Quick delegation template for sub-agent prompts. Use when assigning work to a sub-agent, before invoking the Agent tool, or when preparing prompts for specialized agents. Provides the WHERE-WHAT-WHY framework. For comprehensive delegation guidance, activate the agent-orchestration how-to-delegate skill.
swarm-spawning
Spawn agents and teammates in Claude Code swarms. Use when choosing between subagents vs teammates, selecting agent types (Explore, Plan, general-purpose, plugin agents), configuring spawn backends (in-process, tmux, iterm2), or setting environment variables for spawned agents.
knowledge-explorer
Manage the research/ knowledge base (KB) of tool and library research entries. Use when browsing KB topics, adding new research entries, updating existing entries with dated revisions, fetching GitHub repo metadata into a draft KB entry, or migrating old-format entries to skill-spec frontmatter. Triggers on tasks like "what do we have on X", "add this to the KB", "update the KB entry for Y", "fetch github info for owner/repo", or "migrate old entries".
design-anti-patterns
Enforce anti-AI UI design rules based on the Uncodixfy methodology. Use when generating HTML, CSS, React, Vue, Svelte, or any frontend UI code. Prevents "Codex UI" — the generic AI aesthetic of soft gradients, floating panels, oversized rounded corners, glassmorphism, hero sections in dashboards, and decorative copy. Applies constraints from Linear/Raycast/Stripe/GitHub design philosophy: functional, honest, human-designed interfaces. Triggers on: UI generation, dashboard building, frontend component creation, CSS styling, landing page design, or any task producing visual interface code.
Didn't find tool you were looking for?