diff --git a/graphify/__main__.py b/graphify/__main__.py
index e5ac46c46..3a6fcd2c6 100644
--- a/graphify/__main__.py
+++ b/graphify/__main__.py
@@ -191,6 +191,7 @@ This project has a graphify knowledge graph at graphify-out/.
 Rules:
 - Before answering architecture or codebase questions, read graphify-out/GRAPH_REPORT.md for god nodes and community structure
 - If graphify-out/wiki/index.md exists, navigate it instead of reading raw files
+- For cross-module "how does X relate to Y" questions, prefer `graphify query "<question>"`, `graphify path "<A>" "<B>"`, or `graphify explain "<concept>"` over grep — these traverse the graph's EXTRACTED + INFERRED edges instead of scanning files
 - After modifying code files in this session, run `graphify update .` to keep the graph current (AST-only, no API cost)
 """
 
@@ -206,6 +207,7 @@ This project has a graphify knowledge graph at graphify-out/.
 Rules:
 - Before answering architecture or codebase questions, read graphify-out/GRAPH_REPORT.md for god nodes and community structure
 - If graphify-out/wiki/index.md exists, navigate it instead of reading raw files
+- For cross-module "how does X relate to Y" questions, prefer `graphify query "<question>"`, `graphify path "<A>" "<B>"`, or `graphify explain "<concept>"` over grep — these traverse the graph's EXTRACTED + INFERRED edges instead of scanning files
 - After modifying code files in this session, run `graphify update .` to keep the graph current (AST-only, no API cost)
 """
 
@@ -219,6 +221,7 @@ This project has a graphify knowledge graph at graphify-out/.
 Rules:
 - Before answering architecture or codebase questions, read graphify-out/GRAPH_REPORT.md for god nodes and community structure
 - If graphify-out/wiki/index.md exists, navigate it instead of reading raw files
+- For cross-module "how does X relate to Y" questions, prefer `graphify query "<question>"`, `graphify path "<A>" "<B>"`, or `graphify explain "<concept>"` over grep — these traverse the graph's EXTRACTED + INFERRED edges instead of scanning files
 - After modifying code files in this session, run `graphify update .` to keep the graph current (AST-only, no API cost)
 """
 
@@ -421,6 +424,7 @@ Rules:
 - Before answering architecture or codebase questions, read graphify-out/GRAPH_REPORT.md for god nodes and community structure
 - If graphify-out/wiki/index.md exists, navigate it instead of reading raw files
 - If the graphify MCP server is active, utilize tools like `query_graph`, `get_node`, and `shortest_path` for precise architecture navigation instead of falling back to `grep`
+- If the MCP server is not active, the CLI equivalents are `graphify query "<question>"`, `graphify path "<A>" "<B>"`, and `graphify explain "<concept>"` — prefer these over grep for cross-module questions
 - After modifying code files in this session, run `graphify update .` to keep the graph current (AST-only, no API cost)
 """
 
diff --git a/graphify/report.py b/graphify/report.py
index 82657e1bb..103765dcf 100644
--- a/graphify/report.py
+++ b/graphify/report.py
@@ -49,10 +49,14 @@ def generate(
             "- Verdict: corpus is large enough that graph structure adds value.",
         ]
 
+    from .analyze import _is_file_node as _ifn
+    non_empty = {cid: nodes for cid, nodes in communities.items()
+                 if any(not _ifn(G, n) for n in nodes)}
+
     lines += [
         "",
         "## Summary",
-        f"- {G.number_of_nodes()} nodes · {G.number_of_edges()} edges · {len(communities)} communities detected",
+        f"- {G.number_of_nodes()} nodes · {G.number_of_edges()} edges · {len(non_empty)} communities detected",
         f"- Extraction: {ext_pct}% EXTRACTED · {inf_pct}% INFERRED · {amb_pct}% AMBIGUOUS"
         + (f" · INFERRED: {len(inf_edges)} edges (avg confidence: {inf_avg})" if inf_avg is not None else ""),
         f"- Token cost: {token_cost.get('input', 0):,} input · {token_cost.get('output', 0):,} output",
@@ -60,9 +64,9 @@ def generate(
 
     # Community hub navigation - links to _COMMUNITY_*.md files in the Obsidian vault.
     # Without these, GRAPH_REPORT.md is a dead-end and the vault splits into disconnected components.
-    if communities:
+    if non_empty:
         lines += ["", "## Community Hubs (Navigation)"]
-        for cid in communities:
+        for cid in non_empty:
             label = community_labels.get(cid, f"Community {cid}")
             safe = _safe_community_name(label)
             lines.append(f"- [[_COMMUNITY_{safe}|{label}]]")
@@ -105,7 +109,6 @@ def generate(
             lines.append(f"- **{h.get('label', h.get('id', ''))}** — {node_labels} [{conf_tag}]")
 
     lines += ["", "## Communities"]
-    from .analyze import _is_file_node as _ifn
     for cid, nodes in communities.items():
         label = community_labels.get(cid, f"Community {cid}")
         score = cohesion_scores.get(cid, 0.0)
@@ -141,7 +144,8 @@ def generate(
         if G.degree(n) <= 1 and not _is_file_node(G, n) and not _is_concept_node(G, n)
     ]
     thin_communities = {
-        cid: nodes for cid, nodes in communities.items() if len(nodes) < 3
+        cid: nodes for cid, nodes in communities.items()
+        if 0 < sum(1 for n in nodes if not _is_file_node(G, n)) < 3
     }
     gap_count = len(isolated) + len(thin_communities)
 
diff --git a/pyproject.toml b/pyproject.toml
index 2eece2a6b..0e6c8b6ee 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "graphifyy"
-version = "0.4.24"
+version = "0.4.25"
 description = "AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, Aider, OpenClaw, Factory Droid, Trae, Hermes, Kiro, Google Antigravity) - turn any folder of code, docs, papers, images, or videos into a queryable knowledge graph"
 readme = "README.md"
 license = { file = "LICENSE" }