Generate targeted clarification questions for low-confidence intent routing.
npx skills add https://github.com/reaatech/agent-mesh --skill clarification-reaatech
Please help me install this Agent Skill. Skill: Clarification Source: https://github.com/reaatech/agent-mesh/tree/main/skills/clarification Command: npx skills add https://github.com/reaatech/agent-mesh --skill clarification-reaatech
When the classifier runs with low confidence, this skill generates targeted clarifying questions to refine user intent and improve routing accuracy.
Use the Clarification skill by invoking the generate_clarification_question tool with parameters agent_name, user_input, detected_language, and intent_summary to obtain a clarifying question.
High-intent search queries and answers about installing and using this skill.
Low-confidence intent routing is handled by generating targeted clarifying questions to refine ambiguous user inputs, improving classifier accuracy and directing queries to the correct agent.
To generate clarifying questions, invoke the generate_clarification_question tool with parameters including agent_name, user_input, detected_language, and intent_summary to produce a concise query for the user.
Yes, you can cache clarification results to reduce latency, as the system caches and reuses clarification outputs for identical user inputs and agents to optimize routing performance.
Intent clarification supports localization by providing language-aware fallback questions and structured processing, ensuring robust fallback behavior across multiple detected languages.
Automated clarification requires safe prompts and output validation to prevent prompt injection, and relies on structured toolsets, meaning poorly validated inputs or missing fallback policies can limit its effectiveness.
robit-manRoute natural language requests to matching skills and workflows.
BayramAnnakovDigest unread Telegram messages, extract channel writing styles, and draft posts.
aggelosChatziioannouClassify user queries by intent, complexity, and confidence level.
ShaheerKhawajaRoute natural-language intents to the correct ProductionOS workflow.
