diff --git a/_skill-index.md b/_skill-index.md
index ee1e439..dcfb002 100644
--- a/_skill-index.md
+++ b/_skill-index.md
@@ -47,6 +47,7 @@ description: Master index of all skills in your robot assistant system. Your ass
 | "read this email," "review this email," "open this gmail," "check my gmail" | **gmail-browser** |
 | "make a query," "show all pages tagged," "build a filter," "list tasks in," "show recent notes," "create a silverbullet view," "make a live widget" | **silverbullet-query** |
 | "clipboard history," "from maccy," "what was on my clipboard," "get the last link I copied" | **clipboard-history** |
+| "log food," "add a food item," "what did I eat," "show my food log," "track macros," "food items," "daily food" | **food-tracking** |
 
 ---
 
@@ -250,6 +251,12 @@ description: Master index of all skills in your robot assistant system. Your ass
 **File:** `skills/silverbullet-query/SKILL.md`
 **Dependencies:** SilverBullet running with Space Lua enabled
 
+### Food Tracking
+**Purpose:** Manage food items (catalog with nutrition info) and daily food log entries in MongoDB. Add new foods with macros, search the catalog, log meals and supplements, review eating history.
+**Triggers:** "log food," "add a food item," "what did I eat," "show my food log," "track macros," "food items," "daily food"
+**File:** `skills/food-tracking/SKILL.md`
+**Dependencies:** Python with `pymongo`, `uv` CLI, scripts at `skills/food-tracking/scripts/`
+
 ### Clipboard History
 **Purpose:** Retrieve items from clipboard history using Maccy on macOS. Query the Maccy SQLite database to find recently copied text, URLs, or search for specific content. macOS only.
 **Triggers:** "clipboard history," "from maccy," "what was on my clipboard," "get the last link I copied"
diff --git a/food-tracking/SKILL.md b/food-tracking/SKILL.md
new file mode 100644
index 0000000..095f808
--- /dev/null
+++ b/food-tracking/SKILL.md
@@ -0,0 +1,100 @@
+# Food Tracking
+
+Manage food items and daily food log entries in MongoDB (`lists.food_items`, `lists.food_log`).
+
+## Scripts
+
+Both scripts are standalone Python CLIs. Run with `uv run --with pymongo`:
+
+### food_items.py — `lists.food_items`
+
+Reference item catalog: what foods exist, their nutrition, serving sizes, etc.
+
+```bash
+# Path relative to skill directory
+SCRIPT="skills/food-tracking/scripts/food_items.py"
+
+# Add a new food item
+uv run --with pymongo "$SCRIPT" add \
+  --name "Kefir Oats" \
+  --type food \
+  --serving "1.5 cups" \
+  --description "Whole milk kefir with old-fashioned rolled oats" \
+  --nutrition '{"calories": 228, "protein_g": 12.1, "fat_g": 7.0, "carbs_g": 30.1, "fiber_g": 3.5, "sugar_g": 6.3}'
+
+# Add a medicine
+uv run --with pymongo "$SCRIPT" add \
+  --name "Excedrin Migraine" \
+  --type medicine \
+  --max-dosage "2 caplets"
+
+# Add with arbitrary extra fields via JSON
+uv run --with pymongo "$SCRIPT" add \
+  --name "Chicken Breast" \
+  --type food \
+  --serving "4 oz" \
+  --nutrition '{"calories": 187, "protein_g": 35, "fat_g": 4, "carbs_g": 0}' \
+  --extra '{"recipe_notes": "grilled, no oil"}'
+
+# Get all items
+uv run --with pymongo "$SCRIPT" get
+
+# Filter by type
+uv run --with pymongo "$SCRIPT" get --type food
+
+# Search by name
+uv run --with pymongo "$SCRIPT" get --name "kefir"
+
+# Update an item by name
+uv run --with pymongo "$SCRIPT" update --name "Kefir Oats" --set-serving "2 cups"
+
+# Update an item by _id
+uv run --with pymongo "$SCRIPT" update --id "6a11b0c3..." --set-nutrition '{"calories": 300, "protein_g": 15, "fat_g": 9, "carbs_g": 40}'
+
+# Unset a field
+uv run --with pymongo "$SCRIPT" update --name "Excedrin Migraine" --unset max_dosage
+
+# Merge arbitrary fields
+uv run --with pymongo "$SCRIPT" update --name "Chicken Breast" --set-extra '{"tags": ["dinner", "high-protein"]}'
+```
+
+### food_log.py — `lists.food_log`
+
+Daily consumption log: what was eaten, when, and how much.
+
+```bash
+SCRIPT="skills/food-tracking/scripts/food_log.py"
+
+# Log a food entry
+uv run --with pymongo "$SCRIPT" add --name "Kefir Oats" --type food --amount 1 --unit "serving"
+
+# Log a medicine
+uv run --with pymongo "$SCRIPT" add --name "Excedrin Migraine" --type medicine --amount 1 --unit "caplet"
+
+# Log with extra fields
+uv run --with pymongo "$SCRIPT" add --name "Coffee" --type food --amount 2 --unit "cups" --extra '{"caffeine_mg": 200}'
+
+# Get today's log
+uv run --with pymongo "$SCRIPT" get --today
+
+# Get last 7 days
+uv run --with pymongo "$SCRIPT" get --days 7
+
+# Filter by name or type
+uv run --with pymongo "$SCRIPT" get --name "Excedrin"
+uv run --with pymongo "$SCRIPT" get --type medicine
+
+# Update an entry (requires --id from a previous get)
+uv run --with pymongo "$SCRIPT" update --id "6a11a937..." --set-amount 2 --set-unit "caplets"
+
+# Unset a field
+uv run --with pymongo "$SCRIPT" update --id "6a11a937..." --unset unit
+```
+
+## Workflow Notes
+
+- When logging food, the assistant should run `food_items.py get --name <item>` first to confirm the item exists and show its nutrition info.
+- For new foods not in the catalog, add the item first, then log it.
+- Always present nutrition info back to the user when logging food (pull from `food_items` so the user sees the macros).
+- When searching the internet for nutrition facts, use USDA FoodData Central as the source, then store the result as a food item.
+- The `--extra` flag on both scripts accepts arbitrary JSON, useful for fields not covered by named flags (e.g., recipe details, tags, notes).
diff --git a/food-tracking/scripts/food_items.py b/food-tracking/scripts/food_items.py
new file mode 100644
index 0000000..c97d172
--- /dev/null
+++ b/food-tracking/scripts/food_items.py
@@ -0,0 +1,171 @@
+#!/usr/bin/env python3
+"""CLI for managing food items in MongoDB lists.food_items."""
+
+import argparse
+import json
+import sys
+from datetime import datetime, timezone
+from pymongo import MongoClient
+
+CONN = "mongodb://root:3wwfoUjyk2E2zWELXFlLuHqfw1ALlOp4pb2H5Vq3TImbMIHL2h1u8Jej2mjzCPl@docdb.connorrhodes.com:35563/?tls=true&tlsAllowInvalidCertificates=true"
+
+def get_collection():
+    client = MongoClient(CONN)
+    return client["lists"]["food_items"], client
+
+def cmd_add(args):
+    coll, client = get_collection()
+    doc = {"name": args.name}
+
+    if args.type:
+        doc["type"] = args.type
+    if args.serving:
+        doc["serving"] = args.serving
+    if args.description:
+        doc["description"] = args.description
+    if args.max_dosage:
+        doc["max_dosage"] = args.max_dosage
+    if args.nutrition:
+        try:
+            doc["nutrition"] = json.loads(args.nutrition)
+        except json.JSONDecodeError:
+            print(f"Error: --nutrition must be valid JSON. Got: {args.nutrition}", file=sys.stderr)
+            sys.exit(1)
+    if args.extra:
+        try:
+            extra = json.loads(args.extra)
+            doc.update(extra)
+        except json.JSONDecodeError:
+            print(f"Error: --extra must be valid JSON. Got: {args.extra}", file=sys.stderr)
+            sys.exit(1)
+
+    result = coll.insert_one(doc)
+    out = coll.find_one({"_id": result.inserted_id})
+    out["_id"] = str(out["_id"])
+    print(json.dumps(out, indent=2, default=str))
+    client.close()
+
+def cmd_get(args):
+    coll, client = get_collection()
+    query = {}
+    if args.name:
+        query["name"] = {"$regex": args.name, "$options": "i"}
+    if args.type:
+        query["type"] = {"$regex": args.type, "$options": "i"}
+
+    cursor = coll.find(query).sort("name", 1).limit(args.limit if args.limit else 50)
+    docs = []
+    for d in cursor:
+        d["_id"] = str(d["_id"])
+        docs.append(d)
+
+    if not docs:
+        print("No items found.")
+    else:
+        print(json.dumps(docs, indent=2, default=str))
+    client.close()
+
+def cmd_update(args):
+    coll, client = get_collection()
+
+    # Build filter
+    if args.id:
+        from bson import ObjectId
+        query = {"_id": ObjectId(args.id)}
+    elif args.name:
+        query = {"name": args.name}
+    else:
+        print("Error: provide --id or --name to identify the item", file=sys.stderr)
+        sys.exit(1)
+
+    # Build update
+    update_fields = {}
+    if args.set_name:
+        update_fields["name"] = args.set_name
+    if args.set_type:
+        update_fields["type"] = args.set_type
+    if args.set_serving:
+        update_fields["serving"] = args.set_serving
+    if args.set_description:
+        update_fields["description"] = args.set_description
+    if args.set_max_dosage:
+        update_fields["max_dosage"] = args.set_max_dosage
+    if args.set_nutrition:
+        try:
+            update_fields["nutrition"] = json.loads(args.set_nutrition)
+        except json.JSONDecodeError:
+            print(f"Error: --set-nutrition must be valid JSON", file=sys.stderr)
+            sys.exit(1)
+    if args.set_extra:
+        try:
+            extra = json.loads(args.set_extra)
+            update_fields.update(extra)
+        except json.JSONDecodeError:
+            print(f"Error: --set-extra must be valid JSON", file=sys.stderr)
+            sys.exit(1)
+    if args.unset:
+        for field in args.unset:
+            update_fields[field] = ""
+
+    if not update_fields:
+        print("Error: nothing to update", file=sys.stderr)
+        sys.exit(1)
+
+    set_doc = {"$set": {k: v for k, v in update_fields.items()}}
+    unset_doc = {"$unset": {k: v for k, v in update_fields.items() if v == ""}}
+
+    update = {}
+    if set_doc["$set"]:
+        update["$set"] = set_doc["$set"]
+    if unset_doc["$unset"]:
+        update["$unset"] = unset_doc["$unset"]
+
+    result = coll.update_one(query, update)
+    print(f"Matched: {result.matched_count}, Modified: {result.modified_count}")
+
+    if result.matched_count:
+        updated = coll.find_one(query)
+        updated["_id"] = str(updated["_id"])
+        print(json.dumps(updated, indent=2, default=str))
+    else:
+        print("No matching item found.")
+    client.close()
+
+def main():
+    parser = argparse.ArgumentParser(description="Manage food items in lists.food_items")
+    sub = parser.add_subparsers(dest="command", required=True)
+
+    # add
+    p_add = sub.add_parser("add", help="Add a new food item")
+    p_add.add_argument("--name", required=True)
+    p_add.add_argument("--type")
+    p_add.add_argument("--serving")
+    p_add.add_argument("--description")
+    p_add.add_argument("--max-dosage")
+    p_add.add_argument("--nutrition", help='JSON string, e.g. \'{"calories": 228, "protein_g": 12}\'')
+    p_add.add_argument("--extra", help="JSON string with additional fields to merge into the document")
+
+    # get
+    p_get = sub.add_parser("get", help="Get food items")
+    p_get.add_argument("--name", help="Filter by name (regex)")
+    p_get.add_argument("--type", help="Filter by type (regex)")
+    p_get.add_argument("--limit", type=int, help="Max items to return")
+
+    # update
+    p_upd = sub.add_parser("update", help="Update an existing food item")
+    p_upd.add_argument("--id", help="ObjectId to update")
+    p_upd.add_argument("--name", help="Name to match (if no --id)")
+    p_upd.add_argument("--set-name", dest="set_name")
+    p_upd.add_argument("--set-type", dest="set_type")
+    p_upd.add_argument("--set-serving", dest="set_serving")
+    p_upd.add_argument("--set-description", dest="set_description")
+    p_upd.add_argument("--set-max-dosage", dest="set_max_dosage")
+    p_upd.add_argument("--set-nutrition", dest="set_nutrition", help="JSON string to replace nutrition")
+    p_upd.add_argument("--set-extra", dest="set_extra", help="JSON string with additional fields to set")
+    p_upd.add_argument("--unset", nargs="*", help="Field names to remove")
+
+    args = parser.parse_args()
+    {"add": cmd_add, "get": cmd_get, "update": cmd_update}[args.command](args)
+
+if __name__ == "__main__":
+    main()
diff --git a/food-tracking/scripts/food_log.py b/food-tracking/scripts/food_log.py
new file mode 100644
index 0000000..309ea56
--- /dev/null
+++ b/food-tracking/scripts/food_log.py
@@ -0,0 +1,162 @@
+#!/usr/bin/env python3
+"""CLI for managing food log entries in MongoDB lists.food_log."""
+
+import argparse
+import json
+import sys
+from datetime import datetime, timezone, timedelta
+from pymongo import MongoClient
+
+CONN = "mongodb://root:3wwfoUjyk2E2zWELXFlLuHqfw1ALlOp4pb2H5Vq3TImbMIHL2h1u8Jej2mjzCPl@docdb.connorrhodes.com:35563/?tls=true&tlsAllowInvalidCertificates=true"
+
+def get_collection():
+    client = MongoClient(CONN)
+    return client["lists"]["food_log"], client
+
+def cmd_add(args):
+    coll, client = get_collection()
+    doc = {"timestamp": datetime.now(timezone.utc)}
+
+    if args.name:
+        doc["name"] = args.name
+    if args.type:
+        doc["type"] = args.type
+    if args.amount:
+        doc["amount"] = args.amount
+    if args.unit:
+        doc["unit"] = args.unit
+    if args.extra:
+        try:
+            extra = json.loads(args.extra)
+            doc.update(extra)
+        except json.JSONDecodeError:
+            print(f"Error: --extra must be valid JSON. Got: {args.extra}", file=sys.stderr)
+            sys.exit(1)
+
+    if "name" not in doc:
+        print("Error: --name is required", file=sys.stderr)
+        sys.exit(1)
+
+    result = coll.insert_one(doc)
+    out = coll.find_one({"_id": result.inserted_id})
+    out["_id"] = str(out["_id"])
+    print(json.dumps(out, indent=2, default=str))
+    client.close()
+
+def cmd_get(args):
+    coll, client = get_collection()
+    query = {}
+
+    if args.name:
+        query["name"] = {"$regex": args.name, "$options": "i"}
+    if args.type:
+        query["type"] = {"$regex": args.type, "$options": "i"}
+    if args.today:
+        today_start = datetime.now(timezone.utc).replace(hour=0, minute=0, second=0, microsecond=0)
+        query["timestamp"] = {"$gte": today_start}
+    if args.days:
+        cutoff = datetime.now(timezone.utc) - timedelta(days=args.days)
+        query["timestamp"] = {"$gte": cutoff}
+
+    cursor = coll.find(query).sort("timestamp", -1).limit(args.limit if args.limit else 50)
+    docs = []
+    for d in cursor:
+        d["_id"] = str(d["_id"])
+        docs.append(d)
+
+    if not docs:
+        print("No entries found.")
+    else:
+        print(json.dumps(docs, indent=2, default=str))
+    client.close()
+
+def cmd_update(args):
+    coll, client = get_collection()
+
+    from bson import ObjectId
+
+    if not args.id:
+        print("Error: --id is required for update", file=sys.stderr)
+        sys.exit(1)
+
+    query = {"_id": ObjectId(args.id)}
+
+    update_fields = {}
+    if args.set_name:
+        update_fields["name"] = args.set_name
+    if args.set_type:
+        update_fields["type"] = args.set_type
+    if args.set_amount is not None:
+        update_fields["amount"] = args.set_amount
+    if args.set_unit:
+        update_fields["unit"] = args.set_unit
+    if args.set_extra:
+        try:
+            extra = json.loads(args.set_extra)
+            update_fields.update(extra)
+        except json.JSONDecodeError:
+            print(f"Error: --set-extra must be valid JSON", file=sys.stderr)
+            sys.exit(1)
+    if args.unset:
+        for field in args.unset:
+            update_fields[field] = ""
+
+    if not update_fields:
+        print("Error: nothing to update", file=sys.stderr)
+        sys.exit(1)
+
+    set_doc = {"$set": {k: v for k, v in update_fields.items()}}
+    unset_doc = {"$unset": {k: v for k, v in update_fields.items() if v == ""}}
+
+    update = {}
+    if set_doc["$set"]:
+        update["$set"] = set_doc["$set"]
+    if unset_doc["$unset"]:
+        update["$unset"] = unset_doc["$unset"]
+
+    result = coll.update_one(query, update)
+    print(f"Matched: {result.matched_count}, Modified: {result.modified_count}")
+
+    if result.matched_count:
+        updated = coll.find_one(query)
+        updated["_id"] = str(updated["_id"])
+        print(json.dumps(updated, indent=2, default=str))
+    else:
+        print("No matching entry found.")
+    client.close()
+
+def main():
+    parser = argparse.ArgumentParser(description="Manage food log entries in lists.food_log")
+    sub = parser.add_subparsers(dest="command", required=True)
+
+    # add
+    p_add = sub.add_parser("add", help="Add a food log entry")
+    p_add.add_argument("--name", required=True)
+    p_add.add_argument("--type")
+    p_add.add_argument("--amount", type=float)
+    p_add.add_argument("--unit")
+    p_add.add_argument("--extra", help="JSON string with additional fields")
+
+    # get
+    p_get = sub.add_parser("get", help="Get food log entries")
+    p_get.add_argument("--name", help="Filter by name (regex)")
+    p_get.add_argument("--type", help="Filter by type (regex)")
+    p_get.add_argument("--today", action="store_true", help="Only today's entries")
+    p_get.add_argument("--days", type=int, help="Entries from last N days")
+    p_get.add_argument("--limit", type=int, help="Max entries to return")
+
+    # update
+    p_upd = sub.add_parser("update", help="Update a food log entry")
+    p_upd.add_argument("--id", required=True, help="ObjectId to update")
+    p_upd.add_argument("--set-name", dest="set_name")
+    p_upd.add_argument("--set-type", dest="set_type")
+    p_upd.add_argument("--set-amount", dest="set_amount", type=float)
+    p_upd.add_argument("--set-unit", dest="set_unit")
+    p_upd.add_argument("--set-extra", dest="set_extra", help="JSON string with additional fields to set")
+    p_upd.add_argument("--unset", nargs="*", help="Field names to remove")
+
+    args = parser.parse_args()
+    {"add": cmd_add, "get": cmd_get, "update": cmd_update}[args.command](args)
+
+if __name__ == "__main__":
+    main()