Run a comprehensive health check on an Immich photo library — asset counts, storage usage, metadata completeness, orphaned files, and quality indicators. Use when the user says "library health", "health report", "library status", "library audit", "how healthy is my library", "photo stats", "library overview", "what's in my library", "library report", or any variation of wanting a comprehensive overview of their photo library's state.

Library Health Report

⚠️ Connection Required — ALWAYS CHECK FIRST

Before doing ANYTHING else in this skill, call ping on the Immich MCP server.

If ping succeeds → proceed with the skill normally.
If ping fails or the MCP tools are not available → STOP. Do not continue. Tell the user:

❌ Immich is not connected. This plugin needs a running Immich MCP server to work.

Run /setup-immich-photo-manager to configure your Immich connection. You'll need:

Your Immich server URL (e.g., http://192.168.1.100:2283)

An Immich API key (how to create one)

The MCP server configured (see /setup-immich-photo-manager)

Nothing in this plugin will work until the connection is configured.

Do NOT skip this check. Do NOT try to run any other tool first. Always ping, always block if it fails.

Generate a comprehensive health assessment of an Immich photo library. Covers asset inventory, storage breakdown, metadata quality, and actionable recommendations.

When to Use

First time exploring a new Immich library
Periodic checkups (monthly/quarterly)
After bulk imports to verify everything landed correctly
Before cleanup operations to understand baseline

Report Sections

Section 1: Asset Inventory

Query Immich for the full library snapshot:

-- Total assets by type
SELECT type, count(*) as total,
  pg_size_pretty(sum(("exifInfo"->>'fileSizeInByte')::bigint)) as total_size
FROM asset
WHERE "deletedAt" IS NULL
GROUP BY type;

-- Trash contents
SELECT type, count(*) as trashed
FROM asset
WHERE "deletedAt" IS NOT NULL
GROUP BY type;

Also use the MCP tool get_statistics for the official Immich counts.

Present as:

ASSET INVENTORY
  Photos:          39,596
  Videos:           4,983
  Total live:      44,579
  In trash:         8,433
  Storage used:    ~180 GB

Section 2: Import Sources

Identify where photos came from:

SELECT
  CASE
    WHEN "originalPath" LIKE '%Apple Fotos%' OR "originalPath" LIKE '%Apple Photos%' THEN 'Apple Photos'
    WHEN "originalPath" LIKE '%Google Fotos%' OR "originalPath" LIKE '%Google Photos%' THEN 'Google Photos'
    WHEN "originalPath" LIKE '%upload%' THEN 'Direct Upload'
    ELSE split_part("originalPath", '/', 5)
  END as source,
  count(*) as total,
  min("localDateTime") as earliest,
  max("localDateTime") as latest
FROM asset WHERE "deletedAt" IS NULL
GROUP BY source ORDER BY total DESC;

Section 3: Metadata Completeness

Check for gaps in critical metadata fields:

-- GPS coverage
SELECT
  count(*) as total,
  count(*) FILTER (WHERE "exifInfo"->>'latitude' IS NOT NULL) as has_gps,
  round(100.0 * count(*) FILTER (WHERE "exifInfo"->>'latitude' IS NOT NULL) / count(*), 1) as gps_pct
FROM asset WHERE "deletedAt" IS NULL AND type = 'IMAGE';

-- Date quality
SELECT
  count(*) FILTER (WHERE "exifInfo"->>'dateTimeOriginal' IS NOT NULL) as has_exif_date,
  count(*) FILTER (WHERE extract(hour from "localDateTime") = 12
    AND extract(minute from "localDateTime") = 0) as suspicious_noon,
  count(*) FILTER (WHERE extract(hour from "localDateTime") = 0
    AND extract(minute from "localDateTime") = 0) as suspicious_midnight
FROM asset WHERE "deletedAt" IS NULL AND type = 'IMAGE';

-- Camera info
SELECT
  count(*) FILTER (WHERE "exifInfo"->>'make' IS NOT NULL) as has_camera_make,
  count(*) FILTER (WHERE "exifInfo"->>'lensModel' IS NOT NULL) as has_lens
FROM asset WHERE "deletedAt" IS NULL AND type = 'IMAGE';

Present as:

METADATA QUALITY
  GPS coverage:        72.3% (28,616 of 39,596 photos)
  EXIF dates:          89.1% (35,280 photos)
  Suspicious dates:    1,204 (noon/midnight — likely recovered from paths)
  Camera info:         68.4% (27,072 photos)
  Lens info:           52.1% (20,623 photos)

Section 4: Time Distribution

SELECT extract(year from "localDateTime") as year,
  count(*) as photos,
  count(*) FILTER (WHERE type = 'VIDEO') as videos
FROM asset WHERE "deletedAt" IS NULL
GROUP BY year ORDER BY year;

Highlight years with unusually low counts (potential missing imports).

Section 5: File Format Breakdown

SELECT
  upper(split_part("originalPath", '.', -1)) as format,
  count(*) as total,
  pg_size_pretty(sum(("exifInfo"->>'fileSizeInByte')::bigint)) as size
FROM asset WHERE "deletedAt" IS NULL
GROUP BY format ORDER BY total DESC
LIMIT 15;

Section 6: Recommendations

Based on findings, generate actionable recommendations:

Low GPS coverage? → Suggest metadata-fixer skill
Suspicious dates? → Suggest metadata-fixer skill
Multiple import sources? → Suggest duplicate-report skill
Large trash? → Suggest reviewing and permanently deleting
Year gaps? → Suggest timeline-gaps skill
Heavy storage? → Suggest storage-optimizer skill

Output Format

The report can be presented as:

Inline — formatted text in the conversation
Markdown file — saved to the user's vault/workspace
HTML dashboard — interactive file with charts (uses Chart.js or Recharts)

Always ask the user which format they prefer.

Important Notes

This skill is read-only — it never modifies assets or metadata
Uses both Immich MCP tools (for official counts) and direct database queries (for deep analysis)
Database access requires PostgreSQL connection details
For libraries >50K assets, some queries may take 10-30 seconds
Always show the query execution time so users know what to expect

library-health-report