ApiMesh is an open-source tool that automatically scans codebases to generate OpenAPI 3.0 specifications and an interactive HTML API UI. It targets developers seeking to eliminate manual API documentation, prevent drift, and enable instant API exploration across multiple languages, offering accurate documentation with zero configuration.

How It Works

ApiMesh uses an AI-augmented pipeline: a FileScanner traverses code, heuristics and LLMs detect frameworks (e.g., Express, FastAPI, Rails, Gin), and extract API details. Vector embeddings enrich endpoint context. The process generates a precise swagger.json and embeds it into a self-contained, interactive apimesh-docs.html powered by Swagger UI, requiring no server setup.

Quick Start & Requirements

The recommended installation is via Docker:

docker run --pull always -it --rm -v $(pwd):/workspace qodexai/apimesh:latest

This command initiates an interactive session. The tool supports Python, Node.js/TS, Ruby on Rails, Go, and offers an LLM fallback for other languages. A Docker environment is the primary prerequisite.

Highlighted Details

• Smart Code Discovery: Finds endpoints across frameworks without annotations.
• OpenAPI 3.0 Spec: Generates swagger.json for CI/CD, gateways, and tools.
• Interactive HTML UI: Renders apimesh-docs.html, a self-contained, offline-capable API playground with try-it-out functionality.
• Multi-Language Support: Covers popular frameworks in Python, Node.js, Ruby, and Go, with extensibility.
• Zero-Config Start: Single command for scanning, spec generation, and UI rendering.

Maintenance & Community

Contributions are welcomed via issues and pull requests for enhancing language/framework coverage. The project encourages community involvement to improve automatic API documentation. No specific community channels or notable contributors/sponsors are mentioned.

Licensing & Compatibility

The provided README does not explicitly state the software license. Users should verify licensing terms before adoption, especially for commercial use.

Limitations & Caveats

Reliance on AI and heuristics may lead to edge cases for unsupported or highly customized frameworks. The LLM fallback for unlisted languages might offer reduced reliability compared to native parsers.