claude-code-sub-agent-collective

https://github.com/your-repo/claude-code-collective

[{"step":"Define the task and constraints","action":"Clearly outline the [TASK], [NUMBER] of agents, and [SPECIFIC_SUBTASK] each agent should handle. Include [CONSTRAINTS] like time limits, data formats, or error handling rules.","tip":"Use bullet points for constraints to avoid ambiguity. For example: 'Agent 1 must extract data in JSON format' or 'Agent 2 should retry failed connections 3 times.'"},{"step":"Set up the agent hub","action":"Initialize the claude-code-sub-agent-collective hub in your environment. Ensure all agents have access to required APIs, credentials, and tools (e.g., Playwright for web automation).","tip":"Test agent connectivity before full execution. For Playwright, verify browser compatibility and ensure headless mode is enabled if needed."},{"step":"Execute the coordination","action":"Run the hub-and-spoke model. Monitor agent progress in real-time and log outputs. Use the provided template to structure the coordination request.","tip":"For complex tasks, break the workflow into smaller sub-tasks and assign them to specialized agents (e.g., one for data extraction, another for validation)."},{"step":"Review and resolve issues","action":"Analyze the output summary and errors. Address any unresolved discrepancies or failures manually if the agents cannot resolve them autonomously.","tip":"Prioritize errors based on impact. For example, a single SKU out of sync may be less critical than a systemic API failure."},{"step":"Document and iterate","action":"Save the automation report and update your workflow based on lessons learned. Adjust agent roles or constraints for future runs.","tip":"Store reports in a shared location (e.g., Notion, Google Drive) for team visibility. Use the feedback to refine agent configurations."}]