selitys 3.2.1

A developer onboarding tool that explains backend codebases

Selitys

A tool that explains backend codebases to developers. Point it at a local repo or a GitHub URL and get structured explanations of the architecture, request flow, risks, and more — like a senior engineer onboarding a new teammate.

Available as a CLI, a web UI, or both via Docker.

The name "selitys" is Finnish for "explanation".

🌐 Try It Now

selitys.onrender.com — paste any public GitHub URL and explore the analysis instantly. No install required.

First load may take ~30s if the service is cold (free tier).

---

Screenshots

Overview	Architecture	Dependency Tree

Request Flow	Ask (Q&A)

---

Install

Option 1: pip (CLI only)

pip install selitys

Option 2: Docker (web UI + API)

git clone https://github.com/haxsysgit/selitys.git
cd selitys
cp .env.example .env  # optional: add LLM API key
docker compose up

Open http://localhost:8000 — paste a local path or GitHub URL and go.

Option 3: From source

git clone https://github.com/haxsysgit/selitys.git
cd selitys
pip install -e ".[backend]"

# Start the web UI
cd frontend && npm install && npm run dev &
uvicorn backend.app:app --reload

---

Quick Start

CLI

# Generate markdown docs
selitys explain /path/to/repo -o ./output

# Ask a question (instant, keyword-based)
selitys ask /path/to/repo "what frameworks are used?"

# Ask with LLM for richer answers
selitys ask /path/to/repo "explain the auth flow" --llm

Web UI

Paste any of these into the input field:

/path/to/local/repo
https://github.com/encode/httpx
https://github.com/tiangolo/fastapi

Public GitHub repos are cloned automatically with git clone --depth 1.

---

What You Get

Page	What it shows
Overview	Purpose, tech stack, metrics, domain entities, API surface, risks
Architecture	Subsystems with file trees, patterns, frameworks, language breakdown
Request Flow	Step-by-step request walkthrough with interactive timeline
First Read	Recommended reading order for new developers
Config	Config files, environment variables, entry points
Ask	Q&A — keyword-based or LLM-powered

The CLI generates the same data as markdown files:

File	Description
selitys-overview.md	System purpose, tech stack, domain entities, API surface
selitys-architecture.md	Subsystems, patterns, dependency graph, risk areas
selitys-request-flow.md	Step-by-step request walkthrough
selitys-first-read.md	Recommended reading order
selitys-config.md	Configuration files, environment variables

---

CLI Options

selitys explain

selitys explain /path/to/repo -o ./output

Flag	Description
-o, --output	Output directory (default: ./selitys-output)
--json	Also generate selitys-analysis.json
--watch	Re-run on file changes
--include	Glob patterns to include
--exclude	Glob patterns to exclude
--max-file-size	Skip files larger than this (default: 2MB)

selitys ask

selitys ask /path/to/repo "how does auth work?" --llm

Flag	Env var	Description
--llm	—	Use LLM instead of keyword matching
--api-key	SELITYS_API_KEY	API key for the LLM provider
--base-url	SELITYS_BASE_URL	OpenAI-compatible API base URL
--model	SELITYS_MODEL	Model name (e.g. llama-3.3-70b-versatile)

---

Supported Languages

Language	Frameworks
Python	FastAPI, Flask, Django, SQLAlchemy, Alembic, Pydantic, Celery, pytest
TypeScript / JavaScript	Express, Next.js, NestJS, React, Vue, Angular, Prisma, Sequelize

Python support is the most mature.

---

Who It's For

• New team members joining an unfamiliar codebase
• Solo developers returning to an old project
• Tech leads onboarding new hires
• Reviewers assessing system quality

---

Design Principles

• Read-only — never modifies the analyzed repository
• Repo-specific — tailored explanations, not generic advice
• Evidence-first — claims include file references
• Lightweight — no heavy deps; LLM is optional

---

License

MIT