Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
Codex PPT enables AI agents to automate the creation of PowerPoint presentations by converting text-heavy source materials into structured visual decks. It is designed for automation builders integrating image-based presentation generation into agentic workflows.
The skill processes papers, notes, and reports to produce PPTX files that maintain visual consistency across slides.
WHO IT'S FOR
AI agent / automation builders
integrating image-based presentation generation into agents
Content creators
converting articles and reports into visual slide decks
Technical presenters
creating high-impact visual presentations from technical notes
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Installs: Installs the codex-ppt skill globally for Codex from a GitHub release archive. · Codex
Before you start
For runtime use, an image-generation backend must be available in the current environment: either a built-in tool or a compatible image API.
Download `codex-ppt-skill-v*.zip` from GitHub Releases, extract it, and place the included `codex-ppt` folder at `~/.codex/skills/codex-ppt`. Here `*` denotes the release version in the downloaded filename.
Installs: Links a local codex-ppt repository checkout into Codex's global skills directory for live development and testing. · Codex
Before you start
A local checkout of the codex-ppt-skill repository.
For runtime use, an image-generation backend must be available in the current environment: either a built-in tool or a compatible image API.
In the terminal command, replace `/path/to/codex-ppt-skill` with the absolute path to the local repository checkout, while keeping the destination unchanged.
Installs: Installs the codex-ppt skill for OpenClaw. · OpenClaw
Before you start
python3
For runtime use, an image-generation backend must be available in the current environment: either a built-in tool or a compatible image API. Non-Codex environments usually require configuration of gpt-image-2.5-flare, a third-party image API, or an OpenAI-compatible image-generation endpoint.
Run this command in a terminal.
openclaw skills install codex-ppt
If the OpenClaw skill allowlist is enabled, add `codex-ppt` to that allowlist.
Installs: Installs the codex-ppt skill globally for Claude Code. · Claude Code
Before you start
For runtime use, an image-generation backend must be available in the current environment: either a built-in tool or a compatible image API. Non-Codex environments usually require configuration of gpt-image-2.5-flare, a third-party image API, or an OpenAI-compatible image-generation endpoint.
Installs: Installs the codex-ppt skill globally for Hermes Agent. · Hermes Agent
Before you start
For runtime use, an image-generation backend must be available in the current environment: either a built-in tool or a compatible image API. Non-Codex environments usually require configuration of gpt-image-2.5-flare, a third-party image API, or an OpenAI-compatible image-generation endpoint.
Replace [SOURCE_FILE_PATH] with the path to an article, report, paper, notes file, or outline.
Replace [AUDIENCE] with the intended audience.
Replace [SLIDE_COUNT] with the approximate number of slides desired.
An available built-in image-generation tool or compatible image API will be needed when proceeding to slide generation.
Use the codex-ppt skill to turn [SOURCE_FILE_PATH] into a visually unified image-based presentation for [AUDIENCE]. For the first result, propose a concise outline and visual direction for approximately [SLIDE_COUNT] slides. Respect the skill's approval gates and do not generate the full deck or PPTX yet. I accept that each final slide will be a complete 16:9 image and that individual text boxes, charts, and shapes will not be separately editable.
Claude
Codex
OpenClaw
Hermes