diff --git a/src/skill_seekers/cli/adaptors/base.py b/src/skill_seekers/cli/adaptors/base.py
index 95ecc94..77ea073 100644
--- a/src/skill_seekers/cli/adaptors/base.py
+++ b/src/skill_seekers/cli/adaptors/base.py
@@ -288,7 +288,7 @@ class SkillAdaptor(ABC):
         if not references_dir.exists():
             return
 
-        for ref_file in sorted(references_dir.glob("*.md")):
+        for ref_file in sorted(references_dir.rglob("*.md")):
             if ref_file.is_file() and not ref_file.name.startswith("."):
                 try:
                     content = ref_file.read_text(encoding="utf-8")
@@ -565,7 +565,7 @@ class SkillAdaptor(ABC):
             return ""
 
         toc_lines = []
-        for ref_file in sorted(refs_dir.glob("*.md")):
+        for ref_file in sorted(refs_dir.rglob("*.md")):
             if ref_file.name == "index.md":
                 continue
             title = ref_file.stem.replace("_", " ").title()
diff --git a/src/skill_seekers/cli/adaptors/markdown.py b/src/skill_seekers/cli/adaptors/markdown.py
index f280571..a5961fb 100644
--- a/src/skill_seekers/cli/adaptors/markdown.py
+++ b/src/skill_seekers/cli/adaptors/markdown.py
@@ -261,7 +261,7 @@ Browse the reference files for detailed information on each topic. All files are
         # Add all reference files
         if refs_dir.exists():
             # Sort for consistent ordering
-            ref_files = sorted(refs_dir.glob("*.md"))
+            ref_files = sorted(refs_dir.rglob("*.md"))
 
             for ref_file in ref_files:
                 if ref_file.name == "index.md":
diff --git a/src/skill_seekers/cli/adaptors/openai_compatible.py b/src/skill_seekers/cli/adaptors/openai_compatible.py
index 8f5ab3e..43f66ca 100644
--- a/src/skill_seekers/cli/adaptors/openai_compatible.py
+++ b/src/skill_seekers/cli/adaptors/openai_compatible.py
@@ -218,7 +218,7 @@ Always prioritize accuracy by consulting the attached documentation files before
                 knowledge_dir = temp_path / "knowledge_files"
                 knowledge_count = 0
                 if knowledge_dir.exists():
-                    knowledge_count = len(list(knowledge_dir.glob("*.md")))
+                    knowledge_count = len(list(knowledge_dir.rglob("*.md")))
 
                 client = OpenAI(
                     api_key=api_key,
@@ -364,7 +364,7 @@ Always prioritize accuracy by consulting the attached documentation files before
         references = {}
         total_chars = 0
 
-        for ref_file in sorted(references_dir.glob("*.md")):
+        for ref_file in sorted(references_dir.rglob("*.md")):
             if total_chars >= max_chars:
                 break
 
diff --git a/src/skill_seekers/cli/adaptors/streaming_adaptor.py b/src/skill_seekers/cli/adaptors/streaming_adaptor.py
index 7187f24..80a3a29 100644
--- a/src/skill_seekers/cli/adaptors/streaming_adaptor.py
+++ b/src/skill_seekers/cli/adaptors/streaming_adaptor.py
@@ -197,7 +197,7 @@ class StreamingAdaptorMixin:
         # Reference files
         refs_dir = skill_dir / "references"
         if refs_dir.exists():
-            for ref_file in sorted(refs_dir.glob("*.md")):
+            for ref_file in sorted(refs_dir.rglob("*.md")):
                 if ref_file.is_file() and not ref_file.name.startswith("."):
                     content = ref_file.read_text(encoding="utf-8")
                     char_count = len(content)