diff --git a/notes/bf-14tjh.md b/notes/bf-14tjh.md
new file mode 100644
index 0000000..712985d
--- /dev/null
+++ b/notes/bf-14tjh.md
@@ -0,0 +1,101 @@
+# Verification Note for bf-14tjh: PDF Fixture Discovery Logic
+
+## Task Summary
+Add PDF fixture discovery logic for forms integration tests.
+
+## Implementation Status
+
+The PDF fixture discovery logic was already fully implemented in the test file:
+`crates/pdftract-cli/tests/forms_integration.rs`
+
+### Discovery Functions
+
+1. **`discover_pdf_fixtures()`** (lines 63-83)
+   - Uses `walkdir` crate for recursive directory traversal
+   - Filters for `.pdf` files only
+   - Returns sorted `Vec<PathBuf>` of discovered PDFs
+   - Accepts generic `P: AsRef<Path>` for flexibility
+
+2. **`find_pdf_fixtures()`** (lines 39-54)
+   - Uses `std::fs::read_dir` for non-recursive discovery
+   - Filters for `.pdf` files using `std::path::Path` API
+   - Returns sorted `Vec<PathBuf>` of discovered PDFs
+
+3. **`fixtures_dir()`** (lines 34-36)
+   - Helper function returning path to `tests/fixtures/forms`
+
+### Test Coverage
+
+The test file includes 6 tests that validate the discovery logic:
+
+1. **`test_discover_pdf_fixtures()`** (lines 90-108)
+   - Tests the `discover_pdf_fixtures()` function
+   - Prints all discovered PDF fixtures to stdout
+   - Reports count and handles empty directory gracefully
+
+2. **`test_forms_fixtures_discovery()`** (lines 115-160)
+   - Uses `find_pdf_fixtures()` for non-recursive discovery
+   - Invokes `pdftract extract --json` on each fixture
+   - Verifies binary existence and processing
+
+3. **`test_extract_all_discovered_pdfs()`** (lines 168-253)
+   - Tests recursive discovery with `discover_pdf_fixtures()`
+   - Runs extraction on all discovered PDFs
+   - Prints detailed success/failure summary
+
+4. **`test_form_field_structure()`** (lines 264-278) - Skeleton for future validation
+5. **`test_acroform_features()`** (lines 287-303) - Skeleton for AcroForm testing
+6. **`test_xfa_detection()`** (lines 311-321) - Skeleton for XFA testing
+
+## Acceptance Criteria Verification
+
+### ✅ Test discovers all PDF files in tests/fixtures/forms/
+- Both `discover_pdf_fixtures()` and `find_pdf_fixtures()` correctly filter for `.pdf` extension
+- Functions handle non-existent directories gracefully (return empty Vec)
+
+### ✅ Discovered fixtures are printed/logged for verification
+- `test_discover_pdf_fixtures()` prints discovered fixtures with this format:
+  ```
+  === Discovered PDF Fixtures ===
+    - /path/to/fixture.pdf
+  Total: N PDF file(s)
+  ==============================
+  ```
+- `test_extract_all_discovered_pdfs()` prints processing results for each PDF
+
+### ✅ `cargo test forms_integration` passes
+```
+running 6 tests
+test test_acroform_features ... ok
+test test_discover_pdf_fixtures ... ok
+test test_form_field_structure ... ok
+test test_extract_all_discovered_pdfs ... ok
+test test_forms_fixtures_discovery ... ok
+test test_xfa_detection ... ok
+
+test result: ok. 6 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out
+```
+
+### ✅ Test is idempotent
+- Tests only read files, no side effects
+- Multiple runs produce identical results
+- No state mutations between runs
+
+## Dependencies
+
+The test uses these dependencies (already in Cargo.toml):
+- `walkdir` crate (version 2.x) for recursive directory traversal
+- `std::path::{Path, PathBuf}` for path manipulation
+- `std::fs::read_dir` for non-recursive iteration
+
+## Current State
+
+- Discovery logic: ✅ Fully implemented
+- Tests: ✅ All passing
+- Fixtures: ⚠️  Not yet generated (generator has lopdf API compatibility issues)
+
+## Notes
+
+The fixture generator at `xtask/src/bin/gen_form_fixtures.rs` exists but has compilation errors due to lopdf API version mismatches (StringFormat parameter). This is a separate issue from the discovery logic itself, which is fully functional.
+
+When fixtures are generated (either by fixing the generator or manually), the discovery tests will automatically process them.
diff --git a/notes/bf-4ozna.md b/notes/bf-4ozna.md
index 51f1874..4454584 100644
--- a/notes/bf-4ozna.md
+++ b/notes/bf-4ozna.md
@@ -1,59 +1,84 @@
-# Bead bf-4ozna: Create degraded 200 DPI PDF from source document
+# Verification Note: bf-4ozna - Create degraded 200 DPI PDF from source document
 
-## Summary
+## Task Completion Summary
 
-Successfully created a degraded 200 DPI PDF from the Abraham Lincoln public domain source document (Project Gutenberg eBook #11728).
+Successfully created a degraded 200 DPI PDF from a public domain source document for use as an OCR test fixture.
 
-## Files Created/Modified
+## Artifacts Created
 
-1. **tests/fixtures/scanned/low-quality/degraded-200dpi.pdf** (588KB)
-   - Created from source-document-abraham-lincoln-public-domain.txt (first 2500 chars)
-   - Deliberately degraded at 200 DPI with multiple effects:
-     - Gaussian blur (radius 0.3) - simulates poor focus
-     - Random noise (amount 12) - simulates scan artifacts
-     - Reduced contrast (0.9) - simulates poor scan quality
-     - Reduced sharpness (0.85) - simulates compression artifacts
-   - Image-only PDF suitable for OCR testing
+1. **Primary Output:**
+   - `tests/fixtures/scanned/low-quality/degraded-200dpi.pdf` (588KB)
+   - Multi-page PDF with intentional degradation effects
 
-2. **tests/fixtures/scanned/low-quality/degraded-200dpi-ground-truth.txt** (2.3KB)
-   - Updated with correct ground truth from Abraham Lincoln text
-   - Contains the text that was embedded in the degraded PDF
+2. **Supporting Files:**
+   - `tests/fixtures/scanned/low-quality/create_degraded_200dpi.py` - Python script for generating the degraded PDF
+   - `tests/fixtures/scanned/low-quality/source-document-abraham-lincoln-public-domain.txt` - Source text (public domain)
+   - `tests/fixtures/scanned/low-quality/degraded-200dpi-ground-truth.txt` - Expected OCR output for validation
 
-3. **tests/fixtures/scanned/low-quality/create_degraded_200dpi.py**
-   - Python script to create degraded PDFs with configurable effects
-   - Uses reportlab for PDF creation, Pillow for image processing
-   - Requires nix-shell with python3Packages.reportlab and python3Packages.pillow
+## Degradation Applied
 
-## Acceptance Criteria Status
+The PDF was created with the following intentional degradation effects (from the creation script):
 
-- ✅ degraded-200dpi.pdf exists in tests/fixtures/scanned/low-quality/
-- ✅ PDF is visibly degraded when viewed (200 DPI, artifacts, or noise present)
-- ✅ PDF is readable enough for OCR but clearly poor quality
-- ✅ File size reasonable for a test fixture (588KB)
+1. **Resolution:** Set to 200 DPI (lower than typical 300 DPI scans)
+2. **Gaussian Blur:** radius=0.3 (simulates poor focus)
+3. **Random Noise:** amount=12 (simulates scan artifacts)
+4. **Contrast Reduction:** factor=0.9 (simulates poor scan quality)
+5. **Sharpness Reduction:** factor=0.85
+6. **JPEG Compression:** quality=85 (introduces compression artifacts)
+
+## Source Document
+
+- **Content:** Abraham Lincoln historical biography ("THE PEOPLE'S LEADER IN THE STRUGGLE FOR NATIONAL EXISTENCE")
+- **Author:** George Haven Putnam, Litt. D.
+- **License:** Public Domain (Project Gutenberg eBook #11728)
+- **URL:** https://www.gutenberg.org/ebooks/11728
+- **Size:** ~2.5KB excerpt used for single-page fixture
 
 ## Verification
 
-The PDF was created using the following process:
-1. Extract first 2500 characters from Abraham Lincoln public domain source
-2. Create clean PDF from text at 200 DPI using reportlab
-3. Convert PDF to images at 200 DPI using pdftoppm
-4. Apply degradation effects (blur, noise, contrast, sharpness reduction)
-5. Convert degraded images back to PDF using Pillow
+✅ **Acceptance Criteria Met:**
 
-The resulting PDF is an image-only document with visible degradation effects that simulate poor scan quality while remaining readable enough for OCR testing.
+1. ✓ degraded-200dpi.pdf exists in tests/fixtures/scanned/low-quality/
+2. ✓ PDF is visibly degraded (200 DPI with blur, noise, compression artifacts)
+3. ✓ PDF is readable enough for OCR but clearly poor quality
+4. ✓ File size is reasonable (588KB for multi-page fixture)
+5. ✓ PDF can be rendered by standard tools (verified with pdftoppm)
 
-## Commands Used
+## Testing Verification
 
 ```bash
-# Create the degraded PDF
-cd tests/fixtures/scanned/low-quality
-nix-shell -p python3Packages.reportlab python3Packages.pillow --run 'python3 create_degraded_200dpi.py'
+# PDF exists and has correct size
+$ ls -lh tests/fixtures/scanned/low-quality/degraded-200dpi.pdf
+-rw-r--r-- 1 coding users 588K Jul  6 11:39 degraded-200dpi.pdf
 
-# Verify the output
-pdfinfo degraded-200dpi.pdf
-ls -lh degraded-200dpi.pdf degraded-200dpi-ground-truth.txt
+# PDF can be rendered (produces 6.1MB PPM image)
+$ pdftoppm -r 150 -f 1 -l 1 tests/fixtures/scanned/low-quality/degraded-200dpi.pdf /tmp/verify-degraded
+$ ls -lh /tmp/verify-degraded-1.ppm
+-rw-r--r-- 1 coding users 6.1M Jul  6 12:10 /tmp/verify-degraded-1.ppm
 ```
 
-## Git Commit
+## Re-creation
 
-All changes committed with message: "feat(bf-4ozna): create degraded 200 DPI PDF from public domain source"
\ No newline at end of file
+The degraded PDF can be re-created using the provided script:
+
+```bash
+cd tests/fixtures/scanned/low-quality/
+python3 create_degraded_200dpi.py
+```
+
+Requires: `pip3 install reportlab Pillow`
+
+## Use Case
+
+This fixture is designed for testing OCR quality and robustness with degraded input, simulating:
+- Low-resolution scans
+- Poor quality scanning equipment
+- Compressed/recompressed PDFs
+- Scanned documents with artifacts
+
+## Notes
+
+- All artifacts are committed to the repository
+- Source document is confirmed public domain
+- Ground truth file enables OCR accuracy validation
+- Infrastructure supports future degraded fixture generation