Tweak default built-in document categories

Author: jdav-dev

.devcontainer/devcontainer.json

@@ -14,6 +14,7 @@
         "editor.rulers": [
           98
         ],
+        "git.inputValidation": true,
         "terminal.integrated.defaultProfile.linux": "zsh"
       },
       "extensions": [

lib/archivist/system_calls.ex

@@ -8,17 +8,26 @@ defmodule Archivist.SystemCalls do
   @model "llama3.2"
   @num_ctx 8192
 
-  @categories ~w[
-    identity
-    money
-    medical
-    insurance
-    vehicles
-    legal
-    education
-    manuals
-    miscellaneous
-  ]a
+  @categories [
+    "Vital Records and Identification",
+    "Financial Documents",
+    "Tax Records",
+    "Insurance Documents",
+    "Medical and Health Records",
+    "Property and Real Estate",
+    "Housing and Household",
+    "Vehicle and Transportation",
+    "Employment and Career",
+    "Legal and Estate Planning",
+    "Education and Professional Development",
+    "Family and Household Members",
+    "Warranties and Manuals",
+    "Memberships and Subscriptions",
+    "Travel and Leisure",
+    "Digital Assets and Online Accounts",
+    "Sentimental and Historical",
+    "Miscellaneous and Other"
+  ]
 
   @slug_length 25
 
@@ -27,57 +36,102 @@ defmodule Archivist.SystemCalls do
   extracted from a PDF, and your job is to return the following information in valid JSON format:
 
     - category (string)
-      - Must be exactly one of these: #{Enum.join(@categories, ", ")}.
+      - Must be exactly one of these: #{@categories |> Enum.map(&~s/"#{&1}"/) |> Enum.join(", ")}.
       - This refers to the overall subject area or domain of the document.
       - Below are the category explanations for reference:
-
-        - identity
-          - Personal identification and official records.
-          - Examples: Passport, driver's license, social security card, birth certificate, visas,
-            citizenship papers.
-
-        - money
-          - Banking, taxes, financial records, digital receipts, and home-related financial
-            matters.
-          - Examples: Bank statements, credit card statements, IRS tax returns, W-2s, investment
-            records, mortgage documents, lease agreements, property tax records, home-related
-            bills (utilities, internet, etc.), software receipts.
-
-        - medical
-          - Health records, insurance claims, and medical history.
-          - Examples: Doctor visit summaries, prescriptions, lab results, vaccination records,
-            dental & vision records.
-
-        - insurance
-          - All types of insurance policies and claims.
-          - Examples: Health insurance, auto insurance, home insurance, life insurance, policy
-            renewal notices.
-
-        - vehicles
-          - Documents related to vehicle ownership, maintenance, and insurance.
-          - Examples: Car title, registration, loan documents, repair receipts, maintenance logs,
-            DMV paperwork.
-
-        - legal
-          - Legal documents, employment contracts, estate planning, and home-related legal
-            records.
-          - Examples: Wills, trusts, power of attorney, court records, notarized documents,
-            employment contracts, business ownership documents, home repair contracts, renovation
-            permits, legal agreements related to property.
-
-        - education
-          - Academic records, certifications, and professional development.
-          - Examples: Diplomas, transcripts, student loans, course certificates, professional
-            training records.
-
-        - manuals
-          - Instruction manuals, warranties, and documentation for products you own.
-          - Examples: Appliance manuals, electronics guides, furniture assembly instructions,
-            vehicle owner's manuals.
-
-        - miscellaneous
-          - A catch-all for documents that don't fit any other category.
-          - Examples: Unsorted files, temporary documents, one-off records.
+          - Vital Records and Identification
+            - Description: Documents that establish or verify an individual's identity and
+              significant life events.
+            - Examples: Birth certificates, marriage or divorce certificates, death certificates
+              (for family members), passports, Social Security cards (or equivalents), citizenship
+              or naturalization papers, name change documents.
+          - Financial Documents
+            - Description: Paperwork related to banking, credit, investments, and recurring
+              expenses.
+            - Examples: Bank statements, credit card statements, loan agreements (mortgage,
+              student, car), investment records (stocks, bonds, mutual funds, cryptocurrency),
+              budget worksheets, utility bills, subscription invoices.
+          - Tax Records
+            - Description: All documents needed for tax filing, verification, and historical
+              reference.
+            - Examples: Past tax returns, W-2/1099 forms (or international equivalents), receipts
+              for deductible expenses (charitable donations, medical, business), property tax
+              statements.
+          - Insurance Documents
+            - Description: Policies and claims information for various types of insurance.
+            - Examples: Health insurance policy details, life insurance contracts, auto or
+              homeowners policies, coverage schedules, claim forms, renewal notices.
+          - Medical and Health Records
+            - Description: Personal and family health documentation, including treatments and
+              prescriptions.
+            - Examples: Immunization records, physician or hospital visit summaries, lab test
+              results, prescription information, dental/vision care records, documentation of
+              chronic conditions.
+          - Property and Real Estate
+            - Description: Paperwork detailing real property ownership, transactions, and
+              improvements.
+            - Examples: Mortgage agreements, deeds and titles, closing documents, lease agreements
+              for rental properties, receipts for major renovations, HOA (Homeowners Association)
+              guidelines.
+          - Housing and Household
+            - Description: Day-to-day living documents and service agreements for your home.
+            - Examples: Rental lease agreements (if renting), utility contracts and bills
+              (electricity, water, internet), service or maintenance contracts (e.g., lawn care,
+              pest control), appliance manuals, home repair receipts.
+          - Vehicle and Transportation
+            - Description: Records associated with car ownership, maintenance, and usage.
+            - Examples: Vehicle titles, registration papers, auto insurance policies, maintenance
+              and service records, warranty details, driver's license copies, parking permits.
+          - Employment and Career
+            - Description: Information related to current and past employment, as well as
+              professional growth.
+            - Examples: Employment contracts, offer letters, pay stubs, performance evaluations,
+              benefits guides, separation or termination documents, professional certifications,
+              résumés/CVs.
+          - Legal and Estate Planning
+            - Description: Legally binding papers covering estates, end-of-life directives, and
+                other legal matters.
+            - Examples: Wills, trusts, power of attorney documents, living wills or advance
+                directives, guardianship papers, and court or legal settlement documents.
+          - Education and Professional Development
+            - Description: Records of academic achievements, certifications, and ongoing
+              education.
+            - Examples: Transcripts, diplomas, course certificates, scholarships or grant info,
+              professional licenses, continuing education credits, conference attendance records.
+          - Family and Household Members
+            - Description: Personal documents specific to each household member or dependent.
+            - Examples: Spouse or partner's documents (if kept separately), children's birth
+              certificates, school records, immunization details, childcare arrangements, pet
+              adoption or vaccination papers.
+          - Warranties and Manuals
+            - Description: Documentation for product guarantees and user guides.
+            - Examples: Warranty information for electronics or appliances, user manuals, extended
+              service contracts, purchase receipts for large items or equipment.
+          - Memberships and Subscriptions
+            - Description: Details on recurring membership-based services or organizations.
+            - Examples: Gym memberships, club or association memberships, magazine or streaming
+              subscriptions, loyalty or frequent flyer program statements, renewal notices.
+          - Travel and Leisure
+            - Description: Arrangements and records related to vacations, trips, and leisure
+              activities.
+            - Examples: Travel itineraries, flight tickets, hotel confirmations, visa
+              documentation, travel insurance policies, timeshare contracts, past trip expense
+              receipts.
+          - Digital Assets and Online Accounts
+            - Description: Information and credentials for online identities, cloud services, and
+              digital platforms.
+            - Examples: Password manager references (stored securely), domain registrations, cloud
+              storage subscriptions, digital payment account details (PayPal, etc.), important
+              email or social media account notes.
+          - Sentimental and Historical
+            - Description: Keepsakes and personal or family history items with emotional or
+              genealogical importance.
+            - Examples: Family photos, letters, journals, genealogy research, copies of heirlooms,
+              scrapbooks, memorabilia.
+          - Miscellaneous and Other
+            - Description: A catch-all for documents that do not neatly fit into other categories.
+            - Examples: Personal or hobby-related projects, unusual one-off contracts, event
+              memorabilia, or temporary items awaiting proper classification.
 
     - date (string)
       - Must be a valid ISO 8601 date in the format YYYY-MM-DD.