pdf-official
Process and Fill PDF Forms with Claude
Manually filling PDF forms is tedious and error-prone. This skill enables Claude to extract form fields, validate layouts, and populate PDFs programmatically with accurate data.
スキルZIPをダウンロード
Claudeでアップロード
設定 → 機能 → スキル → スキルをアップロードへ移動
オンにして利用開始
テストする
「pdf-official」を使用しています。 Check fillable fields in employment_application.pdf
期待される結果:
Found 12 fillable fields: first_name (text), last_name (text), email (text), phone (text), address (text), city (text), state (choice), zip (text), position_applied (text), available_date (text), full_time (checkbox), part_time (checkbox). Fields span pages 1-2.
「pdf-official」を使用しています。 Fill the form with: first_name=John, last_name=Doe, email=john@example.com
期待される結果:
Successfully filled 12 fields in employment_application.pdf. Output saved to employment_application_filled.pdf. All field values validated against field types.
「pdf-official」を使用しています。 Create validation image for page 1 of fields.json
期待される結果:
Created validation image at page_1_validation.png with 24 bounding boxes. Red rectangles mark text entry areas, blue rectangles mark labels. No intersections detected.
セキュリティ監査
低リスクStatic analyzer flagged 220 patterns but all are false positives. The 'Ruby backtick' detections are markdown code blocks in documentation, not shell execution. 'Weak crypto' and 'C2 keywords' findings have no basis in actual code. The skill uses legitimate Python PDF libraries (pypdf, pdfplumber, PIL) for expected file operations. No network exfiltration, command injection, or malicious patterns found.
低リスクの問題 (1)
リスク要因
品質スコア
作れるもの
Automated Form Filling for Administrative Work
Office workers can have Claude extract form fields from employment applications, tax forms, or registration documents, then populate them with user-provided information accurately and consistently.
PDF Data Extraction for Research
Researchers can extract tables and structured data from academic PDFs, converting them to CSV or Excel format for analysis without manual copy-pasting.
Document Preparation for Legal or Compliance
Legal teams can merge multiple PDF documents, add watermarks, apply consistent annotations, and fill standardized forms with case-specific information.
これらのプロンプトを試す
Check if this PDF has fillable form fields: [upload PDF]. Run the check_fillable_fields script and tell me what fields are available.
Extract the form field information from [input.pdf] to a JSON file, then fill it with these values: [provide field values]. Create the filled PDF as [output.pdf].
This PDF has no fillable fields. Convert it to images, help me identify bounding boxes for text entry areas, create validation images to verify placement, then add annotations to fill in the form with [data].
I have 50 PDF forms that need the same fields filled with different values from this spreadsheet. Extract the field structure once, then process all PDFs with their corresponding data rows.
ベストプラクティス
- Always run check_fillable_fields first to determine if the PDF has interactive form fields or requires annotation-based filling
- Generate and visually inspect validation images before finalizing non-fillable form annotations to ensure accurate text placement
- Keep original PDFs as backups before running fill operations, as the process creates new output files
回避
- Do not skip validation image inspection - bounding box errors result in text appearing in wrong locations
- Do not assume all PDFs have fillable fields - many scanned forms require manual bounding box identification
- Do not use annotation-based filling for forms that have actual form fields - use the fillable field approach instead