Raccoon PPT Skill is an OpenClaw-integrated remote service for generating entire PowerPoint presentations from a single natural language prompt. It targets users needing rapid, automated content creation, abstracting the underlying PPT generation workflow and providing a seamless user experience by collecting necessary inputs, managing task status, and delivering a final download link.

How It Works

The skill operates by collecting minimal required fields (prompt, role, scene, audience) from user input, either directly or through follow-up questions. It then invokes a PPT OpenAPI to create a job, automatically polling for status updates. It manages task states (e.g., queued, running, waiting_user_input, succeeded) and supports resuming tasks with supplementary details. Internal job IDs are hidden, and the final output is a direct download link, simplifying interaction for the end-user.

Quick Start & Requirements

• Primary Install/Run: Execute Python scripts from the scripts/ directory.
• Prerequisites: Python 3, RACCOON_API_TOKEN environment variable. Optional: RACCOON_API_HOST (defaults to https://xiaohuanxiong.com).
• Example Command: python3 scripts/main.py generate --prompt "AI Agent in enterprise" --role "Trainer" --scene "Training" --audience "Internal"
• Documentation:
  • Skill Definition: SKILL.md
  • API Reference: references/API_REFERENCE.md
  • API Cheatsheet: references/CHEATSHEET.md

Highlighted Details

• Core feature: Generates a complete PPT from a single user sentence.
• Automated task management, including status polling and hiding internal job IDs.
• Supports task resumption and continuation via follow-up prompts.
• Designed for seamless integration within the OpenClaw ecosystem.

Maintenance & Community

The provided README does not contain specific details regarding maintainers, community channels (e.g., Discord, Slack), or project roadmaps.

Licensing & Compatibility

License information is not specified in the README. Compatibility for commercial use or linking with closed-source projects cannot be determined from the provided text.

Limitations & Caveats

The project lacks end-to-end real-environment debugging and implements only a minimal functional loop. Input inference (role, scene, audience) relies on the host model, not complex script parsing. PPT generation tasks take 30 mins-2 hours; recommended interaction involves local task retention and later retrieval, not real-time waiting. Handles server errors for concurrent tasks by prompting retry.