← Back to Skills Marketplace
235
Downloads
0
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install nano-banana-pro-pptx
Description
Generate PowerPoint presentations with AI images using Gemini. Each slide is a full-bleed image. Use for creating visual presentations, slideshows, or image-...
Usage Guidance
This skill largely does what it claims: it builds slide plans, asks Gemini for images/content, and writes images and a .pptx to a local output directory. However: (1) the registry metadata you were shown does not list the GEMINI_API_KEY even though SKILL.md and the script require it — that's an inconsistency worth questioning the publisher about; (2) the skill accepts a GEMINI_BASE_URL (env or arg) which, if pointed at an attacker-controlled endpoint, could expose prompts, slide plans, and image requests — only set that to trusted proxies; (3) the source/homepage is unknown and the owner ID is not human-readable, so prefer to run this in an isolated/sandbox environment (or inspect and run the script locally) before providing your Gemini API key. If you proceed, verify the GEMINI_BASE_URL is an official endpoint (or left unset), consider creating an API key with limited quota, and review the generated slide-plans.json and image-requests.log files after a run.
Capability Analysis
Type: OpenClaw Skill
Name: nano-banana-pro-pptx
Version: 1.0.3
The nano-banana-pro-pptx skill is a legitimate tool designed to generate PowerPoint presentations using the Gemini API for both content outlining and image generation. The Python script (scripts/generate_pptx.py) uses standard libraries like google-genai, python-pptx, and Pillow to create full-bleed image slides based on a user-provided prompt. The instructions in SKILL.md are consistent with the script's functionality, and there is no evidence of data exfiltration, malicious execution, or prompt-injection attacks.
Capability Assessment
Purpose & Capability
Name/description match the code: the script plans slides, requests Gemini image/content models, and assembles PNGs into a .pptx. Declared dependencies in the script header (google-genai, python-pptx, Pillow) are appropriate for the stated purpose. However, the registry metadata provided earlier lists no required environment variables or primary credential while SKILL.md and the script require GEMINI_API_KEY (and optionally GEMINI_BASE_URL); this mismatch is an incoherence to be aware of.
Instruction Scope
Runtime instructions direct the agent to run the bundled script with a prompt and slide count. The script only reads GEMINI_API_KEY and GEMINI_BASE_URL, writes output files (slide-plans.json, images, logs) into the user-specified output directory, and does not attempt to read other system files or unrelated environment variables. It also logs request metadata locally.
Install Mechanism
No install spec is present (instruction-only skill with a bundled script). There are no downloads or archive extracts in the manifest. The script lists expected Python dependencies in comments, which is normal for a Python script.
Credentials
Requiring GEMINI_API_KEY is proportionate to the purpose, but the registry metadata omitted declaring any required env vars while SKILL.md and the script require GEMINI_API_KEY (required) and GEMINI_BASE_URL (optional). Allowing a custom GEMINI_BASE_URL means a user could point requests to an arbitrary endpoint; if that endpoint is untrusted it could capture prompts, plans, and images. The mismatch between metadata and SKILL.md is a notable inconsistency.
Persistence & Privilege
The skill is not always-enabled and does not request system-wide persistence or modify other skills. It writes files only to the output directory it creates. Autonomous invocation is enabled by default but that is the platform norm and not a distinguishing risk here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install nano-banana-pro-pptx - After installation, invoke the skill by name or use
/nano-banana-pro-pptx - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.3
- Updated documentation to clarify how the script selects the API key and base URL, specifying the order of precedence and error behavior if not provided.
- No changes to code—documentation only.
v1.0.2
- Updated script path in usage instructions to reflect new workspace location: ~/.openclaw/workspace/skills/...
- Added interval and retry for image generation requests.
- No functional or feature changes.
v1.0.1
- Added structured metadata for required and optional credentials in the skill documentation.
- No changes to the script or usage instructions.
v1.0.0
Initial release: Generate PowerPoint presentations with full-bleed AI images for each slide, using Gemini for both planning and generation.
- Create visually rich presentations or image-based decks from a single topic prompt.
- Supports custom number of slides (1–50) via `--slides N` (required).
- Adjustable image resolution: choose from 1K, 2K, or 4K.
- Output filename auto-generated or specified via `--filename`.
- Requires Gemini API key (via environment variable or `--api-key`).
- Designed for absolute-path invocation; clear usage and workflow instructions provided.
Metadata
Frequently Asked Questions
What is nano banana pro pptx?
Generate PowerPoint presentations with AI images using Gemini. Each slide is a full-bleed image. Use for creating visual presentations, slideshows, or image-... It is an AI Agent Skill for Claude Code / OpenClaw, with 235 downloads so far.
How do I install nano banana pro pptx?
Run "/install nano-banana-pro-pptx" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is nano banana pro pptx free?
Yes, nano banana pro pptx is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does nano banana pro pptx support?
nano banana pro pptx is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created nano banana pro pptx?
It is built and maintained by Jinpeng (@jinpeng); the current version is v1.0.3.
More Skills