presentation ai - Moses Yebei

Abstract

ALLWEONE® Open source AI presentation generator Gamma Alternative. Create professional slides with customizable themes and AI-generated content in minutes.

The Problem

Teams need polished, branded slide decks quickly, but the standard workflow — outline, copy, design, export — is slow and inconsistent. AI presentation tools lock users into rigid templates or opaque platforms. presentation-ai is an open-source, self-hostable alternative to Gamma.app that generates slides from a prompt and leaves the result fully editable.

What This Does

presentation-ai is a Next.js application that turns a topic prompt into a complete presentation. It uses an AI agent (src/ai/agents/presentation/createAgent.ts) to generate an outline, then slides, with live streaming into an editor. The editor is built on Plate.js and supports rich text, charts, infographics, and custom themes.

The repo contains 1,013 files, nearly all TypeScript/TSX. The core logic lives under src/components/notebook/presentation/editor/, with the AI orchestration in src/ai/. It persists data via Prisma/PostgreSQL and supports multiple LLM backends (OpenAI, Together AI, Ollama, LM Studio).

How It Is Wired

Execution starts at the Next.js API routes in src/app/api/presentation/. The main flow is generate/route.ts → createAgent.ts → LLM tool calls → slide content streamed back to the client. The client-side PresentationGenerationManager.tsx consumes the stream and renders into the Plate editor via EditablePlate.tsx.

The import graph shows two clear hubs with high blast radius:

src/components/notebook/presentation/editor/plugins/chart-plugin.tsx — 62 modules import it; 59 it imports. It sits inside an import cycle.
src/components/notebook/presentation/editor/custom-elements/charts/ChartRenderer.tsx — 61 modules import it, but it only imports 4. It is a stable, widely-depended-on leaf.

Both are part of a 134-module circular dependency cluster. Changing chart-plugin.tsx ripples across dozens of files, and the cycle makes isolated refactoring risky. The editor/lib.tsx (50 importers) and editor/utils (45 importers) are also high-traffic modules.

How To Use It

# Install dependencies
pnpm install

# Set up environment
cp .env.example .env.local
# Fill in DATABASE_URL, auth secrets, and LLM API keys

# Run database migrations
pnpm prisma migrate dev

# Start the dev server
pnpm dev

The README documents these steps. The app requires PostgreSQL and at least one LLM provider key. The .env.example file lists the required variables. There is no Dockerfile, so deployment requires a Node.js host and a managed database.

Real-World Use

A marketing team needs a 10-slide product launch deck. They enter a prompt, pick a theme from the 38 built-in options, and the agent generates an outline. After edits, they export to .pptx or share a public link. The local-models route (src/app/api/presentation/local-models/route.ts) also allows running entirely on-prem with Ollama, which matters for regulated environments.

Code Health & Issues

Static analysis (not opinion) found 251 findings across 989 files: 159 high, 92 medium. The dominant issues:

High - Oversized files — ChartRenderer.tsx is 1,782 lines; editor/lib.tsx and utils/parser.ts are similarly large. Hard to reason about, high change risk.
High - Import cycles — 134 modules are in circular dependencies, including chart-plugin.tsx and ChartRenderer.tsx. Refactoring one requires understanding the whole loop.
High - No tests — 989 source files, zero test files. No CI pipeline exists either, so nothing runs the build on merge.
Medium - Deep nesting — parser.ts and ChartEditorControls.tsx reach indentation depth 6.
Medium - Postinstall script — package.json has a postinstall that fetches a binary; this is a supply-chain risk and should be gated in CI.

The Bottom Line

This is a feature-complete, ambitious product with real depth — the theme system, chart editor, and AI agent are substantial work. The trade-off is maintainability: no tests, no CI, and a tangled import graph make changes risky. Use it if you need a self-hosted Gamma alternative and are willing to invest in test coverage before extending it.

What the analyser found

Deployment readiness

5/7

Container image
CI pipeline
Lockfile committed
Test suite
README
License
No committed secrets

Composition

1,013 files

TSX734
TypeScript251
JSON7
YAML3
JavaScript3
CSS3

ReactNext.jsTailwindLangChain

Module dependencies

The 10 most depended-upon modules of 989, from static import analysis. Red outline marks a module in an import cycle.

graph LR n0__editor_plugins_chart_plugin["chart-plugin"] n1_lements_charts_ChartRenderer["ChartRenderer"] n2_book_presentation_editor_lib["lib"] n3_ok_presentation_editor_utils["utils"] n4_ok_presentation_utils_parser["parser"] n5__components_plate_ui_toolbar["toolbar"] n6_ating_toolbar_ToolbarContext["ToolbarContext"] n7__custom_elements_color_utils["color-utils"] n8_ns_diagram_components_plugin["diagram-components-plugi"] n9_m_elements_presentation_icon["presentation-icon"] n1_lements_charts_ChartRenderer --> n2_book_presentation_editor_lib n1_lements_charts_ChartRenderer --> n0__editor_plugins_chart_plugin n0__editor_plugins_chart_plugin --> n2_book_presentation_editor_lib n8_ns_diagram_components_plugin --> n2_book_presentation_editor_lib n4_ok_presentation_utils_parser --> n2_book_presentation_editor_lib n4_ok_presentation_utils_parser --> n8_ns_diagram_components_plugin classDef cyc stroke:#ef4444,stroke-width:2px; class n0__editor_plugins_chart_plugin,n1_lements_charts_ChartRenderer,n8_ns_diagram_components_plugin cyc;

Findings

MediumNo CI/CD pipeline detected - no automated build/test gate .github/ or CI config

What would stop me shipping this

Ranked by severity × confidence × production reach. Reach is the honest discriminator across a collection that is mostly other people's code: the same finding matters more in something that ships.

high3

medium2

Add a test suite; this repository has nonehigh989 occurrences

989 source files, no test files

Any change ships with no signal that existing behaviour still holds, so a regression reaches production undetected.

Fix: Add one test per public entry point, then a CI step that runs them.

Add a workflow that builds and tests this repositoryhigh

989 source files, no CI configuration

Every change merges with nobody having run the build once, which across this estate is 499 repositories.

Fix: Add a workflow running the project build and test command on push and pull_request.

Add a build gate for the deployable artifacts herehigh4 occurrences

src/components/notebook/presentation/editor/custom-elements/charts/ChartRenderer.tsx

deployment artifacts with no workflow

Something in the repo is meant to reach a server and no automated step has ever validated it, which is the state of 125 repositories here.

Fix: Add a workflow that at least builds the image and validates the manifests.

Enable Dependabot or Renovatemedium

1 manifest(s), no update bot configured

Without a bot a published advisory sits unpatched until someone audits by hand, which across 1,322 repositories means never.

Fix: Commit .github/dependabot.yml covering the repo ecosystems plus github-actions.

Review the install lifecycle script and disable scripts in CImedium

package.json

postinstall

Install scripts are the execution vector for every npm compromise from event-stream onward, and a postinstall that fetches a binary makes the build depend on a URL nobody reviews.

Fix: Move the work into an explicit build step, or set ignore-scripts in CI and run it by name.

Checked deterministically against the repository tree and a bounded set of its files: committed credentials, unpinned actions and base images, missing lockfiles and update bots, workflows that discard failures, published advisories against the declared dependencies, runtime configuration, licensing and notebook reproducibility. No language model is involved in this section.