■ Summariser

What this is

Case Summary turns a GMIAU case-summary export into the canonical gmiau-case-summary/1 JSON that other tools — Letter, Court Doc, Bundle Builder, and Evidence Exhibitor — use to auto-fill client and matter details. Paste the markdown text, drop a Case Summary PDF, or type the details out by hand, and it produces a ready-to-save JSON file.

Typical workflow

1. Open the CMS and export (or copy) the case summary for the client.
2. Go to the 📋 Convert tab and choose your source: Paste text for markdown, Drop PDF for the PDF export, or Type it out to enter the details by hand.
3. Paste the markdown into the text area, or drop one or more Case Summary PDFs onto the drop zone.
4. Click Convert → JSON (text) or wait for the automatic parse (PDF). The Detected fields panel shows what was extracted.
5. Check the detected fields — look for any that read undefined or look wrong, and note them for manual correction if needed.
6. Click ↓ Download JSON to save the file. The filename follows your Config pattern (default: {ref} Case Summary {date}.json).
7. Drop the JSON into ~/.case-summaries/ (or wherever your tools look) so it is available to Letter, Court Doc, and other tools' Import Case Summary pickers.

The tabs

📋 Convert

The main workspace. Choose Paste text to paste a markdown case summary, Drop PDF to drag one or more exported PDFs, or Type it out to fill in a form by hand when there's no export to work from. For PDFs you can batch-process several files at once and download them all in one click. The right panel shows the detected fields and lets you download the JSON or copy it to the clipboard.

⚙️ Config

Set the output filename pattern. The default is {ref} Case Summary {date}.json; you can change it here and it will persist across sessions. The pattern supports {ref} (case reference) and {date} (today's date).

🎨 Settings

App appearance (light / dark / system, theme family) and the app font. These settings sync across every GMIAU Shell tool.

Tips & gotchas

• Markdown format matters. The parser expects the CMS's ### Client / ### Case / ### References section headings. Plain Label: value exports also work.
• PDFs are parsed via pdf.js — the text layer must be present (scanned images without OCR will not extract any fields).
• Batch PDF mode processes each file independently. Download All saves every result in one click; individual files can also be downloaded from the batch list.
• Save the JSON to the right folder. The canonical location is ~/00 Open Cases/.case-summaries/ — that is where Letter and other tools look by default.
• If a field reads blank or wrong, edit it in the consuming tool (Letter, Court Doc) after importing — the converter reflects exactly what the CMS exported.

Questions

Q. The reference field is blank.
A. The CMS export may use a different label spelling. Check the raw JSON (Show raw JSON button) and confirm the field name the parser is looking for matches your CMS version.

Q. Can I drop multiple PDFs at once?
A. Yes — switch to Drop PDF mode, drop several files, and use Download all JSON to get them in one go.

Q. Where should I put the output JSON?
A. Drop it into ~/00 Open Cases/.case-summaries/ (symlinked as 02 Summaries/). Tools that import a case summary scan that folder automatically.

More: the GMIAU Caseworker User Guide in your Obsidian vault covers the full case-summary workflow and which tools consume the JSON. — Confidentiality: this tool runs entirely in your browser; nothing is uploaded. The case reference, client name and all field values are confidential — keep the produced JSON on this machine and handle it through your normal channels.