diff --git a/src/api/recommendations.py b/src/api/recommendations.py
index 49f39bc..35d30c0 100644
--- a/src/api/recommendations.py
+++ b/src/api/recommendations.py
@@ -351,16 +351,19 @@ def generate_pr_body(
 
         body_lines.extend(
             [
-                f"### {description} - {severity}",
+                "<details>",
+                f"<summary><b>{severity} Severity:</b> {description}</summary>",
+                "",
             ]
         )
         if reasoning:
             body_lines.extend(
                 [
-                    "",
                     f"**Rationale:** {reasoning}",
+                    "",
                 ]
             )
+        body_lines.append("</details>")
         body_lines.append("")
 
     body_lines.extend(
diff --git a/src/presentation/github_formatter.py b/src/presentation/github_formatter.py
index 10ad350..e5ed98e 100644
--- a/src/presentation/github_formatter.py
+++ b/src/presentation/github_formatter.py
@@ -23,13 +23,66 @@
 }
 
 
+def _build_collapsible_violations_text(violations: list[Violation]) -> str:
+    """Builds a collapsible Markdown string grouped by severity for a list of violations.
+
+    Args:
+        violations: A list of Violation objects to format.
+
+    Returns:
+        A Markdown formatted string with collapsible details blocks for each severity level.
+    """
+    if not violations:
+        return ""
+
+    text = ""
+    severity_order = ["critical", "high", "medium", "low", "info"]
+    severity_groups: dict[str, list[Violation]] = {s: [] for s in severity_order}
+
+    for violation in violations:
+        sev = violation.severity.value if hasattr(violation.severity, "value") else str(violation.severity)
+        if sev in severity_groups:
+            severity_groups[sev].append(violation)
+        else:
+            severity_groups["low"].append(violation)
+
+    for severity in severity_order:
+        if severity_groups[severity]:
+            emoji = SEVERITY_STR_EMOJI.get(severity, "⚪")
+            count = len(severity_groups[severity])
+
+            text += "<details>\n"
+            text += f"<summary><b>{emoji} {severity.title()} Severity ({count})</b></summary>\n\n"
+
+            for violation in severity_groups[severity]:
+                text += f"### {violation.rule_description or 'Unknown Rule'}\n"
+                text += f"{violation.message}\n"
+                if violation.how_to_fix:
+                    text += f"**How to fix:** {violation.how_to_fix}\n"
+                text += "\n"
+
+            text += "</details>\n\n"
+
+    return text
+
+
 def format_check_run_output(
     violations: list[Violation],
     error: str | None = None,
     repo_full_name: str | None = None,
     installation_id: int | None = None,
 ) -> dict[str, Any]:
-    """Format violations for check run output."""
+    """Format violations for check run output.
+
+    Args:
+        violations: List of rule violations to report.
+        error: Optional error message if rule processing failed entirely.
+        repo_full_name: The full repository name (e.g., owner/repo).
+        installation_id: The GitHub App installation ID.
+
+    Returns:
+        A dictionary matching the GitHub Check Run Output schema containing title, summary, and text.
+    """
     if error:
         # Check if it's a missing rules file error
         if "rules not configured" in error.lower() or "rules file not found" in error.lower():
@@ -75,7 +128,7 @@ def format_check_run_output(
         }
 
     # Group violations by severity
-    severity_order = ["critical", "high", "medium", "low"]
+    severity_order = ["critical", "high", "medium", "low", "info"]
     severity_groups: dict[str, list[Violation]] = {s: [] for s in severity_order}
 
     for violation in violations:
@@ -99,22 +152,9 @@ def format_check_run_output(
 
     # Build detailed text
     text = "# Watchflow Rule Violations\n\n"
-
-    for severity in severity_order:
-        if severity_groups[severity]:
-            emoji = SEVERITY_STR_EMOJI.get(severity, "⚪")
-            text += f"## {emoji} {severity.title()} Severity\n\n"
-
-            for violation in severity_groups[severity]:
-                text += f"### {violation.rule_description or 'Unknown Rule'}\n\n"
-                text += f"{violation.message}\n\n"
-                text += f"Rule validation failed with severity: **{violation.severity}**\n"
-                if violation.how_to_fix:
-                    text += f"**How to fix:** {violation.how_to_fix}\n"
-                text += "\n"
-
+    text += _build_collapsible_violations_text(violations)
     text += "---\n"
-    text += "*To configure rules, edit the `.watchflow/rules.yaml` file in this repository.*"
+    text += "💡 *To configure rules, edit the `.watchflow/rules.yaml` file in this repository.*"
 
     return {"title": f"{len(violations)} rule violations found", "summary": summary, "text": text}
 
@@ -151,35 +191,22 @@ def format_rules_not_configured_comment(
 
 
 def format_violations_comment(violations: list[Violation]) -> str:
-    """Format violations as a GitHub comment."""
-    comment = "## 🚨 Watchflow Rule Violations Detected\n\n"
-
-    # Group violations by severity
-    severity_order = ["critical", "high", "medium", "low"]
-    severity_groups: dict[str, list[Violation]] = {s: [] for s in severity_order}
+    """Format violations as a GitHub comment.
 
-    for violation in violations:
-        sev = violation.severity.value if hasattr(violation.severity, "value") else str(violation.severity)
-        if sev in severity_groups:
-            severity_groups[sev].append(violation)
+    Args:
+        violations: List of rule violations to include in the comment.
 
-    # Add violations by severity (most severe first)
-    for severity in severity_order:
-        if severity_groups[severity]:
-            emoji = SEVERITY_STR_EMOJI.get(severity, "⚪")
-            comment += f"### {emoji} {severity.title()} Severity\n\n"
-
-            for violation in severity_groups[severity]:
-                comment += f"### {violation.rule_description or 'Unknown Rule'}\n\n"
-                comment += f"{violation.message}\n\n"
-                comment += f"Rule validation failed with severity: **{violation.severity}**\n"
-                if violation.how_to_fix:
-                    comment += f"**How to fix:** {violation.how_to_fix}\n"
-                comment += "\n"
+    Returns:
+        A Markdown formatted string suitable for a Pull Request timeline comment.
+        Returns an empty string if there are no violations.
+    """
+    if not violations:
+        return ""
 
+    comment = f"### 🛡️ Watchflow Governance Checks\n**Status:** ❌ {len(violations)} Violations Found\n\n"
+    comment += _build_collapsible_violations_text(violations)
     comment += "---\n"
-    comment += "*This comment was automatically generated by [Watchflow](https://watchflow.dev).*\n"
-    comment += "*To configure rules, edit the `.watchflow/rules.yaml` file in this repository.*"
+    comment += "💡 *Reply with `@watchflow ack [reason]` to override these rules, or `@watchflow help` for commands.*"
 
     return comment
 
@@ -261,7 +288,7 @@ def format_acknowledgment_check_run(
 {format_acknowledgment_summary(acknowledgable_violations, acknowledgments)}
 
 **Violations Requiring Fixes:**
-{format_violations_for_check_run(violations)}
+{_build_collapsible_violations_text(violations)}
 
 Please address the remaining violations or acknowledge them with a valid reason.
 """
diff --git a/tests/unit/presentation/test_github_formatter.py b/tests/unit/presentation/test_github_formatter.py
index 5d10f30..a4c6393 100644
--- a/tests/unit/presentation/test_github_formatter.py
+++ b/tests/unit/presentation/test_github_formatter.py
@@ -16,20 +16,43 @@ def test_format_violations_comment_groups_by_severity():
 
     comment = format_violations_comment(violations)
 
-    assert "## 🚨 Watchflow Rule Violations Detected" in comment
-    assert "### 🔴 Critical Severity" in comment
-    assert "### 🟠 High Severity" in comment
+    assert "### 🛡️ Watchflow Governance Checks" in comment
+    assert "**Status:** ❌ 3 Violations Found" in comment
+    assert "<summary><b>🔴 Critical Severity (1)</b></summary>" in comment
+    assert "<summary><b>🟠 High Severity (2)</b></summary>" in comment
     assert "### Rule 2" in comment
     assert "### Rule 1" in comment
     assert "### Rule 3" in comment
     assert "Fix 1" in comment
     assert "Fix 2" in comment
+    assert "💡 *Reply with `@watchflow ack [reason]`" in comment
 
 
 def test_format_violations_comment_empty():
     comment = format_violations_comment([])
-    assert "## 🚨 Watchflow Rule Violations Detected" in comment
-    assert "---" in comment
+    assert comment == ""
+
+
+def test_build_collapsible_violations_text_fallback_severity():
+    """Test that a violation with an unknown severity falls back to 'low' severity bucket."""
+    # Create a violation and use object.__setattr__ to bypass Pydantic validation for testing
+    v = Violation(rule_description="Weird Rule", severity=Severity.LOW, message="Weird error")
+    object.__setattr__(v, "severity", "super_critical_unknown")
+
+    comment = format_violations_comment([v])
+
+    # It should fall back to low severity
+    assert "<summary><b>🟢 Low Severity (1)</b></summary>" in comment
+    assert "Weird error" in comment
+
+
+def test_build_collapsible_violations_text_info_severity():
+    """Test that INFO severity violations are correctly formatted."""
+    v = Violation(rule_description="Info Rule", severity=Severity.INFO, message="Just an info message")
+    comment = format_violations_comment([v])
+
+    assert "<summary><b>⚪ Info Severity (1)</b></summary>" in comment
+    assert "Just an info message" in comment
 
 
 def test_format_check_run_output_success():
@@ -46,7 +69,7 @@ def test_format_check_run_output_with_violations():
 
     assert "1 rule violations found" in output["title"]
     assert "🚨 1 violations found: 1 high" in output["summary"]
-    assert "## 🟠 High Severity" in output["text"]
+    assert "<summary><b>🟠 High Severity (1)</b></summary>" in output["text"]
     assert "### Missing Issue" in output["text"]