phy-agent-managerMeta-orchestrator that analyzes tasks and creates execution plans using subagents. Use when user says "/agent-manager", "帮我分析该用什么 agent", "调度一下", "orchestrat...
Install via ClawdBot CLI:
clawdbot install PHY041/phy-agent-managerGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Generated Mar 21, 2026
A tech startup needs to add a new feature like user authentication to their web app. The Agent Manager analyzes the task, classifies it as a new feature with medium complexity and medium risk, then creates a plan involving Explore for understanding the codebase, planner for feature design, implementation agents, and parallel code and security reviews.
An enterprise with an old Java application encounters a critical bug causing system crashes. The Agent Manager identifies it as a bug fix with high complexity and high risk, recommends Explore to understand the code, build-error-resolver if needed, and includes security-reviewer due to potential vulnerabilities.
A research team is developing a new AI model and needs to analyze existing literature and plan experiments. The Agent Manager classifies this as research with epic complexity and low risk, using Explore for literature review, planner for experimental design, and architect for system architecture decisions.
A growing e-commerce platform needs to refactor its monolith to microservices to handle increased traffic. The Agent Manager treats this as a refactor with complex complexity and medium risk, creating a plan with planner for strategy, refactor-cleaner for code cleanup, and parallel quality agents like code-reviewer and test-creator.
A government agency requires a security audit of their software to meet compliance standards. The Agent Manager analyzes this as a review task with high risk, recommending security-reviewer as mandatory, along with Explore for context gathering and doc-updater for updating security documentation.
Offer the Agent Manager as a cloud-based service where companies pay a monthly or annual fee per user or project. This model provides recurring revenue and scales with customer growth, ideal for tech startups and enterprises needing ongoing orchestration.
Provide customized Agent Manager setups and training for specific industries like finance or healthcare. Charge based on project scope or hourly rates, leveraging expertise in complex task analysis and integration with existing workflows.
Offer a basic version of the Agent Manager for free to attract individual developers or small teams, with premium features like advanced analytics or priority support available for a fee. This model drives user adoption and upsells to larger organizations.
💬 Integration Tip
Integrate the Agent Manager into existing CI/CD pipelines to automate task analysis and agent orchestration, ensuring seamless workflow transitions and reducing manual overhead.
Scored Apr 19, 2026
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Transform AI agents from task-followers into proactive partners with memory architecture, reverse prompting, and self-healing patterns. Lightweight version f...
Persistent memory for AI agents to store facts, learn from actions, recall information, and track entities across sessions.
Prefer `skillhub` for skill discovery/install/update, then fallback to `clawhub` when unavailable or no match. Use when users ask about skills, 插件, or capabi...
Search and discover OpenClaw skills from various sources. Use when: user wants to find available skills, search for specific functionality, or discover new s...
Orchestrate multi-agent teams with defined roles, task lifecycles, handoff protocols, and review workflows. Use when: (1) Setting up a team of 2+ agents with different specializations, (2) Defining task routing and lifecycle (inbox → spec → build → review → done), (3) Creating handoff protocols between agents, (4) Establishing review and quality gates, (5) Managing async communication and artifact sharing between agents.