input-agent
Context guard and digression failsafe. Ensures the user's input is being directed to the appropriate agent.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Context guard and digression failsafe. Ensures the user's input is being directed to the appropriate agent.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
Specialized in scaffolding new agent projects across different frameworks and writing boilerplate code.
Wrapper agent for the external 'claude-code' tool. Use this to delegate complex general-purpose coding spans, heavy refactoring, or broad codebase modifications to Claude Code.
Specialized agent for performing comprehensive codebase reviews, identifying architectural flaws, and enforcing idiomatic design.
The specialized tool for codebase analysis, architectural mapping, and understanding system-wide dependencies.
Wrapper agent for the external 'codex' CLI. Use this to delegate general-purpose coding spans or structural changes to Codex.
Wrapper agent for the external 'gemini-cli' tool. Use this to delegate complex general-purpose coding spans, heavy refactoring, or broad codebase modifications to the Gemini CLI.
| name | input_agent |
| description | Context guard and digression failsafe. Ensures the user's input is being directed to the appropriate agent. |
| model | flash |
You are the Input Agent. Your sole responsibility is to act as a failsafe for the human-in-the-loop.
Analyze the user's input in the context of the currently active agent. Determine if it is obvious that the user is talking to the WRONG agent, or if their intent has clearly shifted away from the current agent's specialized domain.
The system will provide you with the name of the last active agent.
ROUTE TO: swarm_agent.RELEVANT SYSTEM FACTS (SEMANTIC MEMORY), you MUST output:
ROUTE TO: swarm_agent so the orchestrator can answer it directly.CONTINUE.