obsidian-vault-add

Add a new resource (book, podcast, TV show, movie, person, country, or city) to the vault: create the note, download a photo/cover, update today's daily note (media only), and re-index.

Argument parsing

Arguments: <type> <title> [extra info]

• type (required): book, podcast, show (or tv), movie, person, country (or land), city (or stadt)
• title (required): the resource title
• extra info (optional): author, host, episode number, season/episode, country name (for cities), etc.

Examples:

• /obsidian-vault-add book "Atlas Shrugged"
• /obsidian-vault-add podcast "Aethervox Ehrenfeld"
• /obsidian-vault-add show "Silicon Valley" 1.1
• /obsidian-vault-add movie "Dune"
• /obsidian-vault-add person "Thomas Sowell"
• /obsidian-vault-add country "Japan"
• /obsidian-vault-add city "Kyoto"

If type or title is missing, ask the user before proceeding.

Step-by-step workflow

1. Web search for metadata

Search for the title to gather:

Book: title, subtitle, authors, published year, genres, pages, ISBN, short description (German preferred) Podcast: full name, host(s), language, genres, start year, description (German preferred) TV show: full name, actors, genres, network, seasons, episodes, IMDb rating, description (German preferred) Movie: full name, director(s), actors, genres, year, IMDb rating, description (German preferred) Person: full name, aliases, nationality, birthday (and death if applicable), domain/occupation, Wikipedia URL, notable works (books/articles/videos), short German biography (2–4 sentences) Country: official name, capital, continent, official language(s), currency, aliases (native name, abbreviations), short description (German preferred) City: full name, country, region/state, language(s), aliases (native name), short description (German preferred)

2. Check if note already exists

obsidian read file="<title>" 2>/dev/null | grep -Ev "^(20[0-9]{2}-|Your Obsidian)"

If the note already exists, inform the user and stop (unless they want to update it).

3. Create note from template

obsidian create path="<path>" template="<Template>" 2>/dev/null | grep -Ev "^(20[0-9]{2}-|Your Obsidian)"

4. Fill in frontmatter

Edit the note file directly (Read then Edit) — do NOT use obsidian property:set for list fields as it does not reliably handle arrays. Set all fields in one Edit call by replacing the entire frontmatter block.

IMPORTANT — preserve existing data: When updating an existing note, never discard data already present. Merge new metadata with what's there:

• Keep all existing aliases and append any new ones not already listed
• Keep all existing tags (only remove BOAT if present)
• Keep any fields already filled in; only add/update empty fields
• Keep all existing body content (sections, lists, notes)

Book frontmatter fields: aliases, tags: [Buch], photo, urls, published (YYYY-MM-DD), authors (list), genres (list), rating, pages

Podcast frontmatter fields: aliases, tags: [Podcast], photo, urls, hosts (list), guests, genres (list), language, rating, status, started, ended, episodes

TV show frontmatter fields: aliases, tags: [Serie], photo, urls, actors (list), genres (list), rating, imdb_rating, network, seasons, episodes

Movie frontmatter fields: aliases, tags: [Film], photo, urls, directors (list), actors (list), genres (list), rating, imdb_rating, year

Country frontmatter fields: aliases, tags: [Land], photo, urls, capital, continent, language (list), currency, visited, rating, cities (list of wikilinks to known city notes)

Person frontmatter fields: aliases (other names, pen names, maiden names), tags (Mann/Frau, nationality country, domain tags like Schriftsteller/Philosoph/etc.), photo, urls, related, birthday (YYYY-MM-DD), death (YYYY-MM-DD or empty), location, quotes (list of wikilinks), books (list of wikilinks to vault book notes), articles, podcasts, videos

City frontmatter fields: aliases, tags: [Stadt], photo, urls, country (wikilink), region, language (list), visited, rating

Remove the BOAT tag from tags — notes with links are not BOAT.

5. Add description to body

After the frontmatter, add a short German-language description paragraph (2–4 sentences). Keep any template section headers (## Zusammenfassung, ## Kernaussagen, ## Lieblingsfolgen, etc.).

Person body structure: short German bio paragraph, then ### Fakten (bullet list of key facts), then ### Ideen (empty, for future notes).

The image embed ![[filename.webp]] will be added after cover download (step 6) — do not add it manually.

6. Download cover art

Use inline Python (not the download scripts, which require manual list editing). Use Pillow if available for WebP output, otherwise JPEG.

Book — Open Library by title search:

import json, urllib.request, urllib.parse, io, os

query = "<title>"
url = f"https://openlibrary.org/search.json?title={urllib.parse.quote(query)}&limit=3"
req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0 VaultBot/1.0"})
with urllib.request.urlopen(req, timeout=10) as resp:
    data = json.loads(resp.read())
docs = data.get("docs", [])
cover_id = next((d["cover_i"] for d in docs if d.get("cover_i")), None)
if cover_id:
    img_url = f"https://covers.openlibrary.org/b/id/{cover_id}-L.jpg"
    # download and save as WebP using Pillow

Output path: 99 Meta/assets/books/<title> (<first author>).webp

Podcast — iTunes Search API:

import json, urllib.request, urllib.parse

query = "<podcast name>"
params = urllib.parse.urlencode({"term": query, "entity": "podcast", "limit": 5})
url = f"https://itunes.apple.com/search?{params}"
req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0 VaultBot/1.0"})
with urllib.request.urlopen(req, timeout=10) as resp:
    data = json.loads(resp.read())
artwork = data["results"][0].get("artworkUrl600") if data["results"] else None
# download and save as WebP using Pillow

Output path: 99 Meta/assets/podcasts/<title>.webp

TV show — TVMaze API:

import json, urllib.request, urllib.parse

query = "<show title>"
url = f"https://api.tvmaze.com/singlesearch/shows?q={urllib.parse.quote(query)}&embed=images"
req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0 VaultBot/1.0"})
with urllib.request.urlopen(req, timeout=10) as resp:
    data = json.loads(resp.read())
image_url = data.get("image", {}).get("original") or data.get("image", {}).get("medium")
# download and save as WebP using Pillow

Output path: 99 Meta/assets/series/<title>.webp

Movie — Wikipedia REST API (search for poster):

import json, urllib.request, urllib.parse

title = "<movie title>"
search_url = f"https://en.wikipedia.org/api/rest_v1/page/summary/{urllib.parse.quote(title)}"
req = urllib.request.Request(search_url, headers={"User-Agent": "Mozilla/5.0 VaultBot/1.0"})
with urllib.request.urlopen(req, timeout=10) as resp:
    data = json.loads(resp.read())
image_url = data.get("originalimage", {}).get("source") or data.get("thumbnail", {}).get("source")
# download and save as WebP using Pillow

Output path: 99 Meta/assets/movies/<title>.webp

Country — Wikipedia REST API (flag or representative image):

import json, urllib.request, urllib.parse

title = "<country name>"
search_url = f"https://en.wikipedia.org/api/rest_v1/page/summary/{urllib.parse.quote(title)}"
req = urllib.request.Request(search_url, headers={"User-Agent": "Mozilla/5.0 VaultBot/1.0"})
with urllib.request.urlopen(req, timeout=10) as resp:
    data = json.loads(resp.read())
image_url = data.get("originalimage", {}).get("source") or data.get("thumbnail", {}).get("source")
# download and save as WebP using Pillow

Output path: 99 Meta/assets/countries/<title>.webp

Person — Wikipedia REST API, then resize to max 800px on longest side:

import json, urllib.request, urllib.parse

name = "<person name>"
search_url = f"https://en.wikipedia.org/api/rest_v1/page/summary/{urllib.parse.quote(name)}"
req = urllib.request.Request(search_url, headers={"User-Agent": "Mozilla/5.0 VaultBot/1.0"})
with urllib.request.urlopen(req, timeout=15) as resp:
    data = json.loads(resp.read())
image_url = data.get("originalimage", {}).get("source") or data.get("thumbnail", {}).get("source")
# download, resize to max 800px longest side, save as WebP using Pillow

Output path: 99 Meta/assets/people/<title>.webp

City — Wikipedia REST API:

import json, urllib.request, urllib.parse

title = "<city name>"
search_url = f"https://en.wikipedia.org/api/rest_v1/page/summary/{urllib.parse.quote(title)}"
req = urllib.request.Request(search_url, headers={"User-Agent": "Mozilla/5.0 VaultBot/1.0"})
with urllib.request.urlopen(req, timeout=10) as resp:
    data = json.loads(resp.read())
image_url = data.get("originalimage", {}).get("source") or data.get("thumbnail", {}).get("source")
# download and save as WebP using Pillow

Output path: 99 Meta/assets/cities/<title>.webp

After successful download:

1. Set photo: <filename> in frontmatter (Edit the file)
2. Add ![[<filename>]] to the note body using Edit (insert before the description paragraph) — do NOT use obsidian prepend, as it escapes the ! and produces \![[...]]

If cover download fails, note it to the user and continue without a photo.

Also add the new entry to the appropriate download script so future re-runs include it:

Countries, cities, and persons have no dedicated download script — skip this step for those types.

7. Update today's daily note

Skip this step for `person`, `country`, and `city` — they are reference notes, not media being consumed.

For media types, append to today's daily note using the correct emoji and wikilink format:

obsidian daily:append content="<entry>" 2>/dev/null | grep -Ev "^(20[0-9]{2}-|Your Obsidian)"

If the user provided season/episode or episode title in the arguments, use it; otherwise use a sensible placeholder or ask.

8. Re-index qmd

qmd update && qmd embed

Run this after the note is created and finalized so it's immediately searchable.

9. Report to user

Summarize what was done:

• Note path created
• Metadata filled (list key fields)
• Cover art status (saved path or skip reason)
• Daily note entry added
• qmd indexed

Error handling

• If note creation fails, check if the path already exists
• If cover download returns a tiny image (< 5 KB), skip it and warn the user
• If obsidian daily:append fails, tell the user the entry to add manually
• If qmd embed takes too long, run it in the background with run_in_background: true