From 93bc3fc18a5908d94ce82d7c8fa92ce4b96c0149 Mon Sep 17 00:00:00 2001
From: Alexander Whitestone <alexander@alexanderwhitestone.com>
Date: Wed, 15 Apr 2026 15:06:09 +0000
Subject: [PATCH] fix: add directory exclusions for scan performance (#170)

---
 scripts/automation_opportunity_finder.py | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/scripts/automation_opportunity_finder.py b/scripts/automation_opportunity_finder.py
index dbacc3e..677d943 100644
--- a/scripts/automation_opportunity_finder.py
+++ b/scripts/automation_opportunity_finder.py
@@ -56,6 +56,15 @@ SHELL_COMMAND_PATTERNS = [
     "pip install", "npm install", "cargo build",
 ]
 
+# Directories to skip during scans — large/uninteresting trees
+EXCLUDE_DIRS = frozenset({
+    "node_modules", "venv", ".venv", "__pycache__", ".git",
+    "site-packages", "dist", "build", ".tox", ".mypy_cache",
+    ".pytest_cache", "coverage", ".next", "vendor",
+    "skills",  # hermes skills dir is huge
+    "audio_cache", "skins", "profiles",
+})
+
 # Session tool calls that appear repeatedly — candidates for workflow automation
 TOOL_SEQUENCE_MIN_OCCURRENCES = 3
 
@@ -169,10 +178,11 @@ def analyze_documents(root_dirs: List[str]) -> List[Dict[str, Any]]:
                 continue
             if path.suffix not in doc_extensions:
                 continue
-            # Skip hidden dirs and common non-docs
-            if any(part.startswith(".") for part in path.parts):
+            # Skip excluded dirs and hidden dirs
+            parts = path.relative_to(root).parts if root in path.parents or root == path.parent else path.parts
+            if any(p.startswith(".") or p in EXCLUDE_DIRS for p in parts):
                 continue
-            if "node_modules" in str(path) or "venv" in str(path):
+            if len(parts) > 8:
                 continue
 
             try:
@@ -230,9 +240,10 @@ def analyze_scripts(root_dirs: List[str]) -> List[Dict[str, Any]]:
                 continue
             if path.suffix not in script_extensions:
                 continue
-            if any(part.startswith(".") for part in path.parts):
+            parts = path.relative_to(root).parts if root in path.parents or root == path.parent else path.parts
+            if any(p.startswith(".") or p in EXCLUDE_DIRS for p in parts):
                 continue
-            if "node_modules" in str(path) or "venv" in str(path) or "__pycache__" in str(path):
+            if len(parts) > 8:
                 continue
 
             try: