Getting started

Connect a repo. Merge something. Read the report.

There is nothing to install and no agent to run. Semfora connects through a read-only GitHub App, analyzes in our cloud inside a single-use container, and keeps only names and numbers. The free tier includes monthly analysis and AI allowances, enough to point it at a real repo and decide with your own data.

From connect to first report in one merge

Semfora runs in our cloud. One ephemeral container per analysis: your source is checked out, measured, and deleted when the run ends. What we keep are names and numbers.
1

Connect GitHub

Pick your repos. Semfora gets read-only access through a GitHub App.

semfora.ai/signup
2

Every merge gets analyzed

An ephemeral container maps the code's structure and posts back metrics: names and numbers, never source.

status: INDEXING → READY
3

Ask it anything

The dashboard for your leads, the PR gate for CI, and an AI chat grounded in your own metrics.

chat: “what got riskier this week?”

What happens after you connect

First index

The repo shows INDEXING while the first container maps it, then flips to READY. From then on, every merge to your default branch re-analyzes automatically.

History backfill

Semfora walks your git history commit by commit in the background, so trend charts grow from a single point into the real trajectory of the codebase. You'll see charts fill in over the first hours after connecting.

What we store

Each analysis posts back names and numbers, capped at 128 KB: function and module names, scores, and counts. The cloned source is deleted when the run ends. The data-handling details live in the privacy policy.

AI is optional

The AI chat, summaries, and audits can be turned off per user or across your whole organization in settings. The map, the metrics, and the PR gate work without them.