AI Agent Hub
Back to skills
MindShow AI PPT icon

MindShow AI PPT

Office Efficiency Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @user_0311f249/mindshow-aippt following https://skillhub.cn/install/skillhub.md.

About this skill

What Problem It Solves

MindShow AI PPT targets a concrete workflow: turning a one-line topic or an uploaded document into a downloadable PPT, while letting the caller observe outline generation, content progress, and job status in real time. It is not just a prompt wrapper; it calls AIPPT APIs through a zero-dependency Node.js CLI, using apidoc.json as the source of interface structure so request fields are not hand-guessed.

How It Works

  • API calls: the CLI can list endpoints, inspect schema, call ordinary JSON endpoints, and stream SSE endpoints. Ordinary endpoints print machine-readable JSON; streaming endpoints emit raw data: events.
  • Topic generation: it generates a Markdown outline via /v1/gpt/gen_outline_by_topic; if the user does not specify page count, page_count defaults to 25.
  • Template policy: by default it does not pass template_id, relying on the system-recommended template; it calls /v1/template/list only when the user explicitly asks to choose or view templates, and displays preview_img_list.
  • Content generation: it uses /v1/gpt/gen_content_by_outline to produce full Markdown; the UI should replace content per event and update progress_percent.
  • Job polling: after creating a job with /v1/job/ppt_create_by_markdown, it polls /v1/job/get for status and data.job.progress_percent, then retrieves ppt_url and preview images on success.
  • File generation: it first uploads a document via /v1/resource/upload, then passes the returned resource_id to the outline endpoint and reuses the topic workflow.

Boundaries and Cautions

  • Requires AIPPT_TOKEN; token precedence is --token, environment variable, then ./.env. Do not print the token or store it outside the shell.
  • Check both HTTP status and data.code; when data.code is non-zero, report data.info and request_id.
  • SSE output is not wrapped in outer JSON; callers must parse each data: event.
  • Do not guess request fields; read schema first. The base URL defaults to apidoc.json and should be overridden only when the user explicitly provides another environment.

Use Cases

  • Product editors turn a one-line topic into a 25-page PPT and render outline and content progress live.
  • Consultants upload PDF or DOCX files and generate a downloadable PPT with preview images.
  • Frontend engineers integrate AIPPT APIs and parse SSE events and ordinary JSON success states.
  • Operations display template previews and create PPT jobs with the selected `template_id`.

Best For

  • Product content editors who need to turn topics or uploaded documents into PPTs
  • Frontend engineers integrating AIPPT CLI and parsing SSE streams
  • Indie developers adding a Generate PPT action to their product
  • Consultants presenting template previews and creating PPT jobs for clients