AT Protocol MCP Server

get_notifications

Read-only

Fetch user notifications from AT Protocol (likes, reposts, follows, mentions, replies). Use countOnly to get unread badge count without loading full list.

Instructions

Retrieve notifications from AT Protocol (likes, reposts, follows, mentions, replies). Requires authentication (app password). Use countOnly: true to fetch only the unread badge count cheaply without loading the full list; use mark_notifications_seen to clear the unread state after processing. Subject to per-tool rate limiting.

Input Schema

TableJSON Schema

Name	Required	Description
`limit`	No	Max notifications per page (1–100, default 50).
`cursor`	No	Opaque pagination cursor from the previous response cursor field; omit for the first page.
`seenAt`	No	ISO 8601 timestamp; only return notifications that occurred after this time. Optional filter.
`countOnly`	No	When true, return only the unread count and skip fetching the notification list (cheap badge-number path).

Output Schema

TableJSON Schema

Name	Required	Description
`success`	Yes	Whether the request succeeded.
`unreadCount`	Yes	Number of unread notifications (always present).
`notifications`	No	Notification entries. Present only when countOnly is false/absent.
`cursor`	No	Opaque cursor for the next page. Present only when countOnly is false/absent.
`hasMore`	No	True when a next page is available. Present only when countOnly is false/absent.
`seenAt`	No	The timestamp up to which notifications have been seen. Present only when countOnly is false/absent.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv0.6.2Jun 23, 2026
- removedInput schema / additionalProperties
```
Removed value: -false
```

Changed6 schema fields changedv0.6.0Jun 11, 2026

addedInput schema / properties / countOnly

Added value: +{
+  "description": "When true, return only the unread count and skip fetching the notification list (cheap badge-number path).",
+  "type": "boolean"
+}

addedInput schema / properties / cursor / description

Added value: +"Opaque pagination cursor from the previous response cursor field; omit for the first page."

removedInput schema / properties / limit / default
```
Removed value: -50
```

addedInput schema / properties / limit / description

Added value: +"Max notifications per page (1–100, default 50)."

addedInput schema / properties / seenAt / description

Added value: +"ISO 8601 timestamp; only return notifications that occurred after this time. Optional filter."

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "description": "Notification result. When countOnly is true only unreadCount is present; otherwise the full list is returned.",
+  "properties": {
+    "cursor": {
+      "description": "Opaque cursor for the next page. Present only when countOnly is false/absent.",
+      "type": "string"
+    },
+    "hasMore": {
+      "description": "True when a next page is available. Present only when countOnly is false/absent.",
+      "type": "boolean"
+    },
+    "notifications": {
+      "description": "Notification entries. Present only when countOnly is false/absent.",
+      "items": {
+        "properties": {
+          "author": {
+            "description": "Author of the action that triggered the notification.",
+            "properties": {
+              "avatar": {
+                "type": "string"
+              },
+              "did": {
+                "type": "string"
+              },
+              "displayName": {
+                "type": "string"
+              },
+              "followersCount": {
+                "type": "number"
+              },
+              "followsCount": {
+                "type": "number"
+              },
+              "handle": {
+                "type": "string"
+              },
+              "postsCount": {
+                "type": "number"
+              }
+            },
+            "required": [
+              "did",
+              "handle"
+            ],
+            "type": "object"
+          },
+          "cid": {
+            "description": "CID of the notification record.",
+            "type": "string"
+          },
+          "indexedAt": {
+            "description": "ISO 8601 timestamp when indexed.",
+            "type": "string"
+          },
+          "isRead": {
+            "description": "Whether the notification has been read.",
+            "type": "boolean"
+          },
+          "labels": {
+            "description": "Moderation labels, if any.",
+            "type": "array"
+          },
+          "reason": {
+            "description": "Why this notification was generated. Known values: like, repost, follow, mention, reply, quote, starterpack-joined, verified, unverified, like-via-repost, repost-via-repost, subscribed-post. The AT Protocol lexicon defines this as an open union, so treat unrecognized values as new reason kinds rather than errors.",
+            "type": "string"
+          },
+          "record": {
+            "description": "Raw lexicon record payload.",
+            "type": "object"
+          },
+          "uri": {
+            "description": "AT URI of the notification record.",
+            "type": "string"
+          }
+        },
+        "required": [
+          "uri",
+          "cid",
+          "author",
+          "reason",
+          "isRead",
+          "indexedAt"
+        ],
+        "type": "object"
+      },
+      "type": "array"
+    },
+    "seenAt": {
+      "description": "The timestamp up to which notifications have been seen. Present only when countOnly is false/absent.",
+      "type": "string"
+    },
+    "success": {
+      "description": "Whether the request succeeded.",
+      "type": "boolean"
+    },
+    "unreadCount": {
+      "description": "Number of unread notifications (always present).",
+      "type": "number"
+    }
+  },
+  "required": [
+    "success",
+    "unreadCount"
+  ],
+  "type": "object"
+}

Changed1 schema field changedv0.3.0Jun 9, 2026
- removedInput schema / $schema
```
Removed value: -"http://json-schema.org/draft-07/schema#"
```
First observedv0.2.1Jun 2, 2026

TDQS

A4.7/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond annotations (readOnlyHint, openWorldHint), the description adds value: requires auth, subject to rate limiting, countOnly skips full list. No contradiction with annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three sentences, each purposeful, front-loaded with main action. No wasted words.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a read-only list tool with 4 parameters (all documented), annotations, and output schema, the description covers auth, rate limits, special logic (countOnly), and sibling reference. No gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% with detailed parameter descriptions. The description reinforces countOnly but adds no new parameter semantics. Baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description starts with a clear verb ('Retrieve') and specific resource ('notifications from AT Protocol'), listing types (likes, reposts, etc.). It distinguishes itself from siblings like mark_notifications_seen by explicitly mentioning it.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit guidance: when to use countOnly for cheap badge count, when to use mark_notifications_seen to clear state. Also notes authentication and rate limiting, helping the agent decide context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.