deploy-agents
Deploy multiple AI agents to collaborate or work independently on tasks. Choose parallel, sequential, or fan-out strategies; agents share context to avoid conflicts and redundancy.
Instructions
Deploy multiple AI agents to work on tasks collaboratively or independently. Strategies: 'parallel' (N agents on 1 task), 'sequential' (chained, each sees prior output), 'fan-out' (1 agent per task). Agents communicate via shared context to avoid conflicts and redundancy. Current agent mode: read-only.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| tasks | Yes | Array of task descriptions. Single task = collaborative (parallel/sequential). Multiple tasks = fan-out (one agent per task). | |
| agentCount | No | Number of agents to deploy for parallel/sequential strategies (2-10). Ignored in fan-out mode where agent count = number of tasks. | |
| strategy | No | Execution strategy: 'parallel' = all agents work on 1 task concurrently, 'sequential' = agents chain, each sees prior outputs, 'fan-out' = each agent gets a separate task. | parallel |
| provider | No | CLI provider to use ('claude', 'codex', 'gemini'). Defaults to 'gemini'. | gemini |
| model | No | Optional model override. Defaults to server config. | |
| maxConcurrency | No | Max parallel agent processes (1-10, default 3). Controls resource usage. | |
| context | No | Shared context injected into every agent's prompt. Use for project background, constraints, or coordination instructions. | |
| useAgentTeams | No | Enable Claude Code Agent Teams (Claude provider only). Delegates team coordination to Claude Code's native Agent Teams feature. Requires CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS enabled in Claude Code settings. See: https://code.claude.com/docs/en/agent-teams |