# The Genome > How Empryo scans your repo into a ranked dependency graph (PageRank, git co-change, 37 languages) and injects the map into every prompt. On startup, the Genome engine scans your codebase and builds a graph - every file, symbol, and import edge - then ranks it. The rendered result, injected at the top of every prompt, is the **Genome**: what the agent actually sees. The agent never wastes a turn grepping for "where does auth live" - it already sees the map. ## What you see The Genome appears in every prompt, looking like this: ``` src/services/auth/session.ts (→12) +SessionManager Tracks active sessions, refresh tokens, expiry +SessionStore ``` - `+` - exported symbol - `(→12)` - 12 files depend on this one - `[NEW]` - file changed recently - One-line summary after the name (when semantic summaries are on) ## What makes it useful - **Ranked by PageRank** - files imported by lots of others rank higher. - **Personalized per turn** - files you just edited or read get boosted. - **Git-aware** - files that always change together get pulled in too. - **Real-time** - edits re-index immediately. - **37 languages** - TypeScript, Python, Rust, Go, Java, Ruby, C, C++, Swift, Kotlin, and more. ## Config Everything runs out of the box. If you want to tune it: ```json { "genome": true, "semanticSummaries": "synthetic" } ``` | Field | Values | |-------|--------| | `genome` | `true` (default) or `false` to disable | | `semanticSummaries` | `"synthetic"` (fast, code-derived, default), `"ast"`, `"llm"` (one-line LLM descriptions), `"off"` | `/Genome` opens the settings panel (the `genome` config keys below tune the engine that builds the map). ## What gets indexed The scan sees what git sees: tracked and untracked files, minus anything `.gitignore` hides. Dot-directories (`.github`, `.claude`, `.repro-*`) and build output (`node_modules`, `dist`, `build`, `target`, `coverage`, …) are pruned even when tracked. Outside a git repository the same rules run over a plain directory walk. To keep something out of the map that git should still track (fixture repos, benchmark inputs, a brand kit of SVGs), add a **`.empryoignore`** at the project root. Same syntax as `.gitignore`, including negations and `**`: ```gitignore # Indexer-only excludes — git keeps these, the Genome never sees them testenvs/ bench/ assets/ apps/webapp/public/ ``` What it changes: - The file drops out of the map, the symbol graph, PageRank, and the explore/`code_script` file listing (`api.files`, `api.grep`). - `read` and `grep` on an explicit path still work. It is hidden from discovery, not from the agent. - It takes effect on an already-indexed repo: saving `.empryoignore` (or `.gitignore`) triggers a rescan, and the newly excluded rows are pruned. Everything else is skipped on the mtime/size gate, so the rescan costs seconds, not a fresh index. If more than 80% of the index would vanish in one go the prune is refused as a listing error, `/genome` → clear forces it. Large repositories are capped at 10 000 indexed files, ranked by git recency. The cap is announced in `/genome`. Trimming noise with `.empryoignore` is how you make sure the cap lands on fixtures, not source. ## Skip the scan For quick one-shot questions, skip the startup scan: ```bash empryo --headless --no-genome "what's the version?" ``` Or set `EMPRYO_NO_GENOME=1`.