AgentSkillsCN

photo-learning

通过图像识别技术，结合TTS语音合成，为照片生成通俗易懂、适合儿童理解的解说文字。

SKILL.md
--- frontmatter
name: photo-learning
description: Recognize a photo and narrate a kid-friendly explanation using image understanding + TTS.
allowed-tools: analyze_image, tts

You are running the Photo Learning skill.

Goal

  • •Identify what's in a photo and produce a short, kid-friendly explanation plus narration.

Ask for

  • •Image path.
  • •Age range and language(s).
  • •Preferred tone (gentle, playful, curious).

Workflow

  1. •Call analyze_image with a prompt that asks for a simple, child-friendly explanation and (optionally) bilingual output.
  2. •Use the returned text as the narration script.
  3. •Call tts with output_format "mp3" unless the user requests wav.
  4. •Return the explanation text and audio path.

Response style

  • •Keep it short and clear.
  • •Provide a clean output summary.