diff --git a/.github/workflows/daily-safe-output-optimizer.md b/.github/workflows/daily-safe-output-optimizer.md
index 71801355bb5..e4d2e2fc616 100644
--- a/.github/workflows/daily-safe-output-optimizer.md
+++ b/.github/workflows/daily-safe-output-optimizer.md
@@ -367,6 +367,8 @@ After updating the tool description:
 - **Be evidence-based**: Show actual error examples, not assumptions
 - **Be actionable**: Recommend specific description improvements
 
+- **Report Formatting**: Use h3 (###) or lower for all headers in your report to maintain proper document hierarchy. Wrap long sections in `<details><summary>Section Name</summary>` tags to improve readability and reduce scrolling.
+
 ### Issue Creation Rules
 
 - **Skip if no tool description issues found**: Don't create issue for workflow prompt issues only
diff --git a/.github/workflows/design-decision-gate.md b/.github/workflows/design-decision-gate.md
index e3b48b7bb5f..62470c3cb26 100644
--- a/.github/workflows/design-decision-gate.md
+++ b/.github/workflows/design-decision-gate.md
@@ -337,6 +337,10 @@ All ADRs are stored in `docs/adr/` as Markdown files numbered by PR number (e.g.
 > 🔒 *This PR cannot merge until an ADR is linked in the PR body.*
 ```
 
+### Report Formatting
+
+- **Report Formatting**: Use h3 (###) or lower for all headers in your report to maintain proper document hierarchy. Wrap long sections in `<details><summary>Section Name</summary>` tags to improve readability and reduce scrolling.
+
 ## Step 4b: If ADR Found — Verify Implementation Matches
 
 If an ADR **is** found (either in the PR body, on the PR branch, or in a linked issue), verify that the implementation aligns with the stated decision.
diff --git a/.github/workflows/spec-librarian.md b/.github/workflows/spec-librarian.md
index d7726a7b4d5..0c6c829e621 100644
--- a/.github/workflows/spec-librarian.md
+++ b/.github/workflows/spec-librarian.md
@@ -328,9 +328,10 @@ The following specifications are outdated:
 1. **Be thorough**: Check ALL packages, not just a sample
 2. **Be precise**: Reference exact file paths, function names, and dates
 3. **Be actionable**: Every finding should have a clear recommendation
-4. **Use progressive disclosure**: Wrap details in `<details>` tags
-5. **One issue per run**: The `max: 1` limit ensures no issue spam
-6. **Skip if open**: The `skip-if-match` rule prevents duplicate issues
+4. **Report Formatting**: Use h3 (###) or lower for all headers in your report to maintain proper document hierarchy. Wrap long sections in `<details><summary>Section Name</summary>` tags to improve readability and reduce scrolling.
+5. **Use progressive disclosure**: Wrap details in `<details>` tags
+6. **One issue per run**: The `max: 1` limit ensures no issue spam
+7. **Skip if open**: The `skip-if-match` rule prevents duplicate issues
 
 ## Success Criteria
 
diff --git a/.github/workflows/test-quality-sentinel.md b/.github/workflows/test-quality-sentinel.md
index cb608ba09c9..0adfa2da340 100644
--- a/.github/workflows/test-quality-sentinel.md
+++ b/.github/workflows/test-quality-sentinel.md
@@ -451,6 +451,9 @@ After posting the comment, submit a pull request review based on the verdict:
 
 ## Guidelines
 
+### Report Formatting
+- **Report Formatting**: Use h3 (###) or lower for all headers in your report to maintain proper document hierarchy. Wrap long sections in `<details><summary>Section Name</summary>` tags to improve readability and reduce scrolling.
+
 ### Analysis Scope
 - **Focus only on new and changed tests** — do not analyze unchanged test files
 - **Support Go (`*_test.go`) and JavaScript (`*.test.cjs`, `*.test.js`)** as primary targets; note other languages but don't score them