Git’s power hides in a single, invisible folder called .git. When you run git init, the directory materializes, and every subsequent command writes into it. Inside, HEAD points to the current branch, while config stores project‑specific settings.

The heart of the system lives in objects/, where Git stores immutable blobs (file contents), trees (directory structures), and commits (snapshots). Each object is named by a cryptographic hash—usually SHA‑1 or SHA‑256—ensuring data integrity and deduplication. Branches, tags, and remote pointers live under refs/, and their movement history is recorded in logs/, giving you a safety net for recovery.

The index acts as a staging buffer, letting you pick exactly what goes into the next commit. Optional scripts in hooks/ can automate checks before a commit or after a push. Understanding this architecture turns Git from a black box into a predictable, debuggable tool, and explains why commands like git add and git commit feel so natural.