diff --git a/.claude/skills/lexicons/SKILL.md b/.claude/skills/lexicons/SKILL.md
new file mode 100644
index 00000000..3d04e88b
--- /dev/null
+++ b/.claude/skills/lexicons/SKILL.md
@@ -0,0 +1,304 @@
+---
+name: lexicons
+description: Reference for the AT Protocol lexicon system. Use when working with lexicons, adding new record types, or modifying AT Protocol schemas.
+user-invocable: false
+---
+
+# Lexicon System
+
+## Overview
+
+Lexicons define the schema for AT Protocol records. This project has two namespaces:
+- **`pub.leaflet.*`** - Leaflet-specific lexicons (documents, publications, blocks, etc.)
+- **`site.standard.*`** - Standard site lexicons for interoperability
+
+The lexicons are defined as TypeScript in `lexicons/src/`, built to JSON in `lexicons/pub/leaflet/` and `lexicons/site/standard/`, and TypeScript types are generated in `lexicons/api/`.
+
+## Key Files
+
+- **`lexicons/src/*.ts`** - Source definitions for `pub.leaflet.*` lexicons
+- **`lexicons/site/standard/**/*.json`** - JSON definitions for `site.standard.*` lexicons (manually maintained)
+- **`lexicons/build.ts`** - Builds TypeScript sources to JSON
+- **`lexicons/api/`** - Generated TypeScript types and client
+- **`package.json`** - Contains `lexgen` script
+
+## Running Lexicon Generation
+
+```bash
+npm run lexgen
+```
+
+This runs:
+1. `tsx ./lexicons/build.ts` - Builds `pub.leaflet.*` JSON from TypeScript
+2. `lex gen-api` - Generates TypeScript types from all JSON lexicons
+3. `tsx ./lexicons/fix-extensions.ts` - Fixes import extensions
+
+## Adding a New pub.leaflet Lexicon
+
+### 1. Create the Source Definition
+
+Create a file in `lexicons/src/` (e.g., `lexicons/src/myLexicon.ts`):
+
+```typescript
+import { LexiconDoc } from "@atproto/lexicon";
+
+export const PubLeafletMyLexicon: LexiconDoc = {
+  lexicon: 1,
+  id: "pub.leaflet.myLexicon",
+  defs: {
+    main: {
+      type: "record",  // or "object" for non-record types
+      key: "tid",
+      record: {
+        type: "object",
+        required: ["field1"],
+        properties: {
+          field1: { type: "string", maxLength: 1000 },
+          field2: { type: "integer", minimum: 0 },
+          optionalRef: { type: "ref", ref: "other.lexicon#def" },
+        },
+      },
+    },
+    // Additional defs for sub-objects
+    subType: {
+      type: "object",
+      properties: {
+        nested: { type: "string" },
+      },
+    },
+  },
+};
+```
+
+### 2. Add to Build
+
+Update `lexicons/build.ts`:
+
+```typescript
+import { PubLeafletMyLexicon } from "./src/myLexicon";
+
+const lexicons = [
+  // ... existing lexicons
+  PubLeafletMyLexicon,
+];
+```
+
+### 3. Update lexgen Command (if needed)
+
+If your lexicon is at the top level of `pub/leaflet/` (not in a subdirectory), add it to the `lexgen` script in `package.json`:
+
+```json
+"lexgen": "tsx ./lexicons/build.ts && lex gen-api ./lexicons/api ./lexicons/pub/leaflet/document.json ./lexicons/pub/leaflet/myLexicon.json ./lexicons/pub/leaflet/*/* ..."
+```
+
+Note: Files in subdirectories (`pub/leaflet/*/*`) are automatically included.
+
+### 4. Add to authFullPermissions (for record types)
+
+If your lexicon is a record type that users should be able to create/update/delete, add it to the `authFullPermissions` permission set in `lexicons/src/authFullPermissions.ts`:
+
+```typescript
+import { PubLeafletMyLexicon } from "./myLexicon";
+
+// In the permissions collection array:
+collection: [
+  // ... existing lexicons
+  PubLeafletMyLexicon.id,
+],
+```
+
+### 5. Regenerate Types
+
+```bash
+npm run lexgen
+```
+
+### 6. Use the Generated Types
+
+```typescript
+import { PubLeafletMyLexicon } from "lexicons/api";
+
+// Type for the record
+type MyRecord = PubLeafletMyLexicon.Record;
+
+// Validation
+const result = PubLeafletMyLexicon.validateRecord(data);
+if (result.success) {
+  // result.value is typed
+}
+
+// Type guard
+if (PubLeafletMyLexicon.isRecord(data)) {
+  // data is typed as Record
+}
+```
+
+## Adding a New site.standard Lexicon
+
+### 1. Create the JSON Definition
+
+Create a file in `lexicons/site/standard/` (e.g., `lexicons/site/standard/myType.json`):
+
+```json
+{
+  "lexicon": 1,
+  "id": "site.standard.myType",
+  "defs": {
+    "main": {
+      "type": "record",
+      "key": "tid",
+      "record": {
+        "type": "object",
+        "required": ["field1"],
+        "properties": {
+          "field1": {
+            "type": "string",
+            "maxLength": 1000
+          }
+        }
+      }
+    }
+  }
+}
+```
+
+### 2. Regenerate Types
+
+```bash
+npm run lexgen
+```
+
+The `site/*/* site/*/*/*` globs in the lexgen command automatically pick up new files.
+
+## Common Lexicon Patterns
+
+### Referencing Other Lexicons
+
+```typescript
+// Reference another lexicon's main def
+{ type: "ref", ref: "pub.leaflet.publication" }
+
+// Reference a specific def within a lexicon
+{ type: "ref", ref: "pub.leaflet.publication#theme" }
+
+// Reference within the same lexicon
+{ type: "ref", ref: "#myDef" }
+```
+
+### Union Types
+
+```typescript
+{
+  type: "union",
+  refs: [
+    "pub.leaflet.pages.linearDocument",
+    "pub.leaflet.pages.canvas",
+  ],
+}
+
+// Open union (allows unknown types)
+{
+  type: "union",
+  closed: false,  // default is true
+  refs: ["pub.leaflet.content"],
+}
+```
+
+### Blob Types (for images/files)
+
+```typescript
+{
+  type: "blob",
+  accept: ["image/*"],  // or specific types like ["image/png", "image/jpeg"]
+  maxSize: 1000000,  // bytes
+}
+```
+
+### Color Types
+
+The project has color types defined:
+- `pub.leaflet.theme.color#rgb` / `#rgba`
+- `site.standard.theme.color#rgb` / `#rgba`
+
+```typescript
+// In lexicons/src/theme.ts
+export const ColorUnion = {
+  type: "union",
+  refs: [
+    "pub.leaflet.theme.color#rgba",
+    "pub.leaflet.theme.color#rgb",
+  ],
+};
+```
+
+## Normalization Between Formats
+
+Use `lexicons/src/normalize.ts` to convert between `pub.leaflet` and `site.standard` formats:
+
+```typescript
+import {
+  normalizeDocument,
+  normalizePublication,
+  isLeafletDocument,
+  isStandardDocument,
+  getDocumentPages,
+} from "lexicons/src/normalize";
+
+// Normalize a document from either format
+const normalized = normalizeDocument(record);
+if (normalized) {
+  // normalized is always in site.standard.document format
+  console.log(normalized.title, normalized.site);
+
+  // Get pages if content is pub.leaflet.content
+  const pages = getDocumentPages(normalized);
+}
+
+// Normalize a publication
+const pub = normalizePublication(record);
+if (pub) {
+  console.log(pub.name, pub.url);
+}
+```
+
+## Handling in Appview (Firehose Consumer)
+
+When processing records from the firehose in `appview/index.ts`:
+
+```typescript
+import { ids } from "lexicons/api/lexicons";
+import { PubLeafletMyLexicon } from "lexicons/api";
+
+// In filterCollections:
+filterCollections: [
+  ids.PubLeafletMyLexicon,
+  // ...
+],
+
+// In handleEvent:
+if (evt.collection === ids.PubLeafletMyLexicon) {
+  if (evt.event === "create" || evt.event === "update") {
+    let record = PubLeafletMyLexicon.validateRecord(evt.record);
+    if (!record.success) return;
+
+    // Store in database
+    await supabase.from("my_table").upsert({
+      uri: evt.uri.toString(),
+      data: record.value as Json,
+    });
+  }
+  if (evt.event === "delete") {
+    await supabase.from("my_table").delete().eq("uri", evt.uri.toString());
+  }
+}
+```
+
+## Publishing Lexicons
+
+To publish lexicons to an AT Protocol PDS:
+
+```bash
+npm run publish-lexicons
+```
+
+This runs `lexicons/publish.ts` which publishes lexicons to the configured PDS.
diff --git a/.claude/skills/notifications/SKILL.md b/.claude/skills/notifications/SKILL.md
new file mode 100644
index 00000000..908953f5
--- /dev/null
+++ b/.claude/skills/notifications/SKILL.md
@@ -0,0 +1,150 @@
+---
+name: notifications
+description: Reference for the notification system. Use when adding new notification types or modifying notification handling.
+user-invocable: false
+---
+
+# Notification System
+
+## Overview
+
+Notifications are stored in the database and hydrated with related data before being rendered. The system supports multiple notification types (comments, subscriptions, etc.) that are processed in parallel.
+
+## Key Files
+
+- **`src/notifications.ts`** - Core notification types and hydration logic
+- **`app/(home-pages)/notifications/NotificationList.tsx`** - Renders all notification types
+- **`app/(home-pages)/notifications/Notification.tsx`** - Base notification component
+- Individual notification components (e.g., `CommentNotification.tsx`, `FollowNotification.tsx`)
+
+## Adding a New Notification Type
+
+### 1. Update Notification Data Types (`src/notifications.ts`)
+
+Add your type to the `NotificationData` union:
+
+```typescript
+export type NotificationData =
+  | { type: "comment"; comment_uri: string; parent_uri?: string }
+  | { type: "subscribe"; subscription_uri: string }
+  | { type: "your_type"; your_field: string }; // Add here
+```
+
+Add to the `HydratedNotification` union:
+
+```typescript
+export type HydratedNotification =
+  | HydratedCommentNotification
+  | HydratedSubscribeNotification
+  | HydratedYourNotification; // Add here
+```
+
+### 2. Create Hydration Function (`src/notifications.ts`)
+
+```typescript
+export type HydratedYourNotification = Awaited<
+  ReturnType<typeof hydrateYourNotifications>
+>[0];
+
+async function hydrateYourNotifications(notifications: NotificationRow[]) {
+  const yourNotifications = notifications.filter(
+    (n): n is NotificationRow & { data: ExtractNotificationType<"your_type"> } =>
+      (n.data as NotificationData)?.type === "your_type",
+  );
+
+  if (yourNotifications.length === 0) return [];
+
+  // Fetch related data with joins
+  const { data } = await supabaseServerClient
+    .from("your_table")
+    .select("*, related_table(*)")
+    .in("uri", yourNotifications.map((n) => n.data.your_field));
+
+  return yourNotifications.map((notification) => ({
+    id: notification.id,
+    recipient: notification.recipient,
+    created_at: notification.created_at,
+    type: "your_type" as const,
+    your_field: notification.data.your_field,
+    yourData: data?.find((d) => d.uri === notification.data.your_field)!,
+  }));
+}
+```
+
+Add to `hydrateNotifications` parallel array:
+
+```typescript
+const [commentNotifications, subscribeNotifications, yourNotifications] = await Promise.all([
+  hydrateCommentNotifications(notifications),
+  hydrateSubscribeNotifications(notifications),
+  hydrateYourNotifications(notifications), // Add here
+]);
+
+const allHydrated = [...commentNotifications, ...subscribeNotifications, ...yourNotifications];
+```
+
+### 3. Trigger the Notification (in your action file)
+
+```typescript
+import { Notification, pingIdentityToUpdateNotification } from "src/notifications";
+import { v7 } from "uuid";
+
+// When the event occurs:
+const recipient = /* determine who should receive it */;
+if (recipient !== currentUser) {
+  const notification: Notification = {
+    id: v7(),
+    recipient,
+    data: {
+      type: "your_type",
+      your_field: "value",
+    },
+  };
+  await supabaseServerClient.from("notifications").insert(notification);
+  await pingIdentityToUpdateNotification(recipient);
+}
+```
+
+### 4. Create Notification Component
+
+Create a new component (e.g., `YourNotification.tsx`):
+
+```typescript
+import { HydratedYourNotification } from "src/notifications";
+import { Notification } from "./Notification";
+
+export const YourNotification = (props: HydratedYourNotification) => {
+  // Extract data from props.yourData
+
+  return (
+    <Notification
+      timestamp={props.created_at}
+      href={/* link to relevant page */}
+      icon={/* icon or avatar */}
+      actionText={<>Message to display</>}
+      content={/* optional additional content */}
+    />
+  );
+};
+```
+
+### 5. Update NotificationList (`NotificationList.tsx`)
+
+Import and render your notification type:
+
+```typescript
+import { YourNotification } from "./YourNotification";
+
+// In the map function:
+if (n.type === "your_type") {
+  return <YourNotification key={n.id} {...n} />;
+}
+```
+
+## Example: Subscribe Notifications
+
+See the implementation in:
+- `src/notifications.ts:88-125` - Hydration logic
+- `app/lish/subscribeToPublication.ts:55-68` - Trigger
+- `app/(home-pages)/notifications/FollowNotification.tsx` - Component
+- `app/(home-pages)/notifications/NotificationList.tsx:40-42` - Rendering
diff --git a/.claude/skills/spec/SKILL.md b/.claude/skills/spec/SKILL.md
new file mode 100644
index 00000000..dbc201e9
--- /dev/null
+++ b/.claude/skills/spec/SKILL.md
@@ -0,0 +1,85 @@
+---
+name: spec
+description: Draft design documents and specs with research-informed questioning
+disable-model-invocation: true
+---
+
+# Spec Writing
+
+## When to Use
+
+Use this skill when the user explicitly requests a spec, design document, or similar planning artifact. Do not auto-trigger for routine tasks.
+
+## Process
+
+### 1. Initial Scoping
+
+When starting a spec:
+- Ask the user for a brief description of what they want to build/change
+- Reason about which question categories matter for this particular work—derive these from the problem, not a checklist
+
+### 2. Interrogation Loop
+
+Conduct open-ended questioning until the designer signals completion:
+
+**Research-informed questions**: Before asking about an area, do just-in-time codebase research. Find relevant files, understand current patterns, identify constraints. Reference specific files and functions in questions when it adds clarity.
+
+**Adaptive scope**: When the designer marks something as out of scope, update your mental model and don't revisit. However, flag when you believe there are gaps that could cause problems:
+- "This approach assumes X—is that intentional?"
+- "The current implementation of Y in `src/foo.ts:42` handles Z differently. Should we align or diverge?"
+
+**Question style**:
+- Ask one focused question at a time, or a small related cluster
+- Ground questions in what you've learned from the codebase
+- Avoid hypotheticals that don't apply to this codebase
+- When presenting options, describe tradeoffs without recommending unless asked
+
+### 3. Spec Writing
+
+When the designer indicates readiness, produce the spec document.
+
+## Output
+
+Save to: `/specs/YYYY-MM-DD-short-name.md`
+
+### Required Elements
+
+**Title and status** (draft | approved | implemented)
+
+**Goal**: What this achieves and why. 1-3 sentences.
+
+**Design**: The substance of what will be built/changed. For each significant component or concern:
+- Describe the approach
+- State key decisions with their rationale inline
+- Reference specific files, functions, and types where relevant
+- Structure subsections to fit the problem—no fixed format
+
+**Implementation**: Ordered steps that an agent or developer can execute. Each step should:
+- Be concrete and actionable
+- Reference specific files/functions to modify or create
+- Be sequenced correctly (dependencies before dependents)
+
+### Optional
+
+- Background context (only if necessary for understanding)
+- Open questions (only if unresolved items remain)
+
+Add other sections if the problem demands it.
+
+## Writing Style
+
+- No filler, hedging, or preamble
+- No "This document describes..." or "In this spec we will..."
+- Start sections with substance, not meta-commentary
+- Use precise technical language
+- Keep decisions and rationale tight—one sentence each when possible
+- Code references use `file/path.ts:lineNumber` or `functionName` in backticks
+- Prefer concrete over abstract; specific over general
+
+## Status Lifecycle
+
+- **draft**: Work in progress. Should not be merged to main.
+- **approved**: Designer has signed off. Ready for implementation.
+- **implemented**: Work is complete. Spec is now historical record.
+
+Update status in the document as it progresses. Specs are point-in-time snapshots—do not update content after implementation begins except to change status.