diff --git a/backend/app/scanners/entropy.py b/backend/app/scanners/entropy.py
index 8cfda41..4c7080c 100644
--- a/backend/app/scanners/entropy.py
+++ b/backend/app/scanners/entropy.py
@@ -44,6 +44,15 @@ def is_allowlisted(token: str) -> bool:
 
 
 def run_entropy(repo_dir: Path) -> List[Finding]:
+    """
+    Scan the repository directory for files containing high entropy strings.
+
+    Args:
+        repo_dir: Path to the directory to scan.
+
+    Returns:
+        A list of Finding objects discovered during the scan.
+    """
     findings: List[Finding] = []
     ignored_extensions = {
         ".png",
@@ -62,65 +71,78 @@ def run_entropy(repo_dir: Path) -> List[Finding]:
         ".eot",
         ".ttf",
     }
+    ignored_dirs = {
+        "node_modules",
+        "venv",
+        ".venv",
+        "build",
+        "dist",
+        "target",
+        ".next",
+        ".cache",
+        "coverage",
+        "vendor",
+        "__pycache__",
+        ".git",
+    }
+
+    MAX_FILE_SIZE = 1 * 1024 * 1024
 
     for file_path in repo_dir.rglob("*"):
-        if (
-            not file_path.is_file()
-            or file_path.suffix.lower() in ignored_extensions
-            or ".git" in file_path.parts
-        ):
+        if any(part in ignored_dirs for part in file_path.parts):
+            continue
+
+        if not file_path.is_file() or file_path.suffix.lower() in ignored_extensions:
+            continue
+
+        if file_path.stat().st_size > MAX_FILE_SIZE:
             continue
 
         try:
-            content = file_path.read_text(encoding="utf-8", errors="ignore")
+            with open(file_path, "r", encoding="utf-8", errors="ignore") as f:
+                for line_idx, line in enumerate(f, start=1):
+                    for match in STRING_LITERAL_REGEX.finditer(line):
+                        token = match.group(2)
+
+                        if is_allowlisted(token):
+                            continue
+
+                        entropy_score = calculate_shannon_entropy(token)
+                        if entropy_score > 4.0:
+                            relative_path = str(file_path.relative_to(repo_dir))
+                            finding_id = f"entropy:high-entropy-string:{relative_path}:{line_idx}"
+                            severity = "HIGH"
+
+                            raw_data = {
+                                "id": finding_id,
+                                "severity": severity,
+                                "location": {"path": relative_path},
+                                "metadata": {"cwe_category": "CWE-798"},
+                            }
+                            ml_features = extract_features(
+                                raw_data, scanner_name="entropy"
+                            )
+
+                            findings.append(
+                                Finding(
+                                    id=finding_id,
+                                    category="secret",
+                                    severity=severity,
+                                    title="High Entropy String Detected",
+                                    description=f"Potential hardcoded secret discovered (Entropy: {entropy_score:.2f})",
+                                    location=Location(
+                                        path=relative_path,
+                                        start_line=line_idx,
+                                        end_line=line_idx,
+                                    ),
+                                    metadata={
+                                        "engine": "entropy",
+                                        "entropy_score": entropy_score,
+                                        "token_preview": token[:10],
+                                    },
+                                    features=ml_features,
+                                )
+                            )
         except Exception:
             continue
-
-        lines = content.splitlines()
-        for line_idx, line in enumerate(lines, start=1):
-            for match in STRING_LITERAL_REGEX.finditer(line):
-                token = match.group(2)
-
-                if is_allowlisted(token):
-                    continue
-
-                entropy_score = calculate_shannon_entropy(token)
-                if entropy_score > 4.0:
-                    relative_path = str(file_path.relative_to(repo_dir))
-                    finding_id = (
-                        f"entropy:high-entropy-string:{relative_path}:{line_idx}"
-                    )
-                    severity = "HIGH"
-
-                    raw_data_for_extractor = {
-                        "id": finding_id,
-                        "severity": severity,
-                        "location": {"path": relative_path},
-                        "metadata": {"cwe_category": "CWE-798"},
-                    }
-
-                    ml_features = extract_features(
-                        raw_data_for_extractor, scanner_name="entropy"
-                    )
-
-                    findings.append(
-                        Finding(
-                            id=finding_id,
-                            category="secret",
-                            severity=severity,
-                            title="High Entropy String Detected",
-                            description=f"Potential hardcoded secret or token discovered (Entropy: {entropy_score:.2f})",
-                            location=Location(
-                                path=relative_path,
-                                start_line=line_idx,
-                                end_line=line_idx,
-                            ),
-                            metadata={
-                                "engine": "entropy",
-                                "entropy_score": entropy_score,
-                                "token_preview": token[:10],
-                            },
-                            features=ml_features,
-                        )
-                    )
     return findings