diff --git a/ingestion/internal/api/handler.go b/ingestion/internal/api/handler.go
index 0637cde..141ca0a 100644
--- a/ingestion/internal/api/handler.go
+++ b/ingestion/internal/api/handler.go
@@ -11,6 +11,7 @@ import (
 	"strings"
 	"time"
 
+	"github.com/mathiasbq/hyperguild/ingestion/internal/extract"
 	"github.com/mathiasbq/hyperguild/ingestion/internal/pipeline"
 	"github.com/mathiasbq/hyperguild/ingestion/internal/search"
 )
@@ -214,16 +215,16 @@ func (h *Handler) IngestPath(w http.ResponseWriter, r *http.Request) {
 			if !supportedExtensions[ext] {
 				return nil
 			}
-			content, readErr := os.ReadFile(path)
+			content, readErr := extract.Text(path)
 			if readErr != nil {
-				allWarnings = append(allWarnings, fmt.Sprintf("read %s: %v", path, readErr))
+				allWarnings = append(allWarnings, fmt.Sprintf("extract %s: %v", path, readErr))
 				return nil
 			}
 			source := req.Source
 			if source == "" {
 				source = filepath.Base(path)
 			}
-			result, runErr := pipeline.Run(r.Context(), h.pipeline, h.brainDir, string(content), source, req.DryRun)
+			result, runErr := pipeline.Run(r.Context(), h.pipeline, h.brainDir, content, source, req.DryRun)
 			if runErr != nil {
 				allWarnings = append(allWarnings, fmt.Sprintf("ingest %s: %v", path, runErr))
 				return nil
@@ -243,16 +244,16 @@ func (h *Handler) IngestPath(w http.ResponseWriter, r *http.Request) {
 			writeError(w, http.StatusBadRequest, fmt.Sprintf("unsupported file extension: %s", ext))
 			return
 		}
-		content, readErr := os.ReadFile(req.Path)
+		content, readErr := extract.Text(req.Path)
 		if readErr != nil {
-			writeError(w, http.StatusInternalServerError, fmt.Sprintf("read file: %v", readErr))
+			writeError(w, http.StatusInternalServerError, fmt.Sprintf("extract text: %v", readErr))
 			return
 		}
 		source := req.Source
 		if source == "" {
 			source = filepath.Base(req.Path)
 		}
-		result, runErr := pipeline.Run(r.Context(), h.pipeline, h.brainDir, string(content), source, req.DryRun)
+		result, runErr := pipeline.Run(r.Context(), h.pipeline, h.brainDir, content, source, req.DryRun)
 		if runErr != nil {
 			h.logger.Error("ingest-path failed", "path", req.Path, "err", runErr)
 			writeError(w, http.StatusInternalServerError, "ingest error")
diff --git a/ingestion/internal/watcher/watcher.go b/ingestion/internal/watcher/watcher.go
index a65db79..2a1c0fb 100644
--- a/ingestion/internal/watcher/watcher.go
+++ b/ingestion/internal/watcher/watcher.go
@@ -11,6 +11,7 @@ import (
 	"time"
 	"unicode"
 
+	"github.com/mathiasbq/hyperguild/ingestion/internal/extract"
 	"github.com/mathiasbq/hyperguild/ingestion/internal/pipeline"
 )
 
@@ -88,12 +89,12 @@ func processFile(ctx context.Context, cfg Config, path, date string) error {
 	filename := filepath.Base(path)
 	source := deriveSource(filename)
 
-	content, err := os.ReadFile(path)
+	content, err := extract.Text(path)
 	if err != nil {
-		return fmt.Errorf("read file: %w", err)
+		return fmt.Errorf("extract text: %w", err)
 	}
 
-	_, runErr := pipeline.Run(ctx, cfg.Pipeline, cfg.BrainDir, string(content), source, false)
+	_, runErr := pipeline.Run(ctx, cfg.Pipeline, cfg.BrainDir, content, source, false)
 	if runErr != nil {
 		// Move to failed/.
 		failedDir := filepath.Join(cfg.BrainDir, "raw", "failed")