Narrareach

List direct messages

list_conversations

Read-onlyIdempotent

Read the writer's stored direct-message conversations from Instagram, Facebook, TikTok, Bluesky and Substack, newest message first. Read-only: it does not fetch from the platforms or send anything. Conversations are read from the platforms every few minutes (Substack about once an hour); check sync.lastSyncedAt for freshness. state=inbox (the default) lists conversations where the reader wrote last within the past 7 days; history lists the rest; all lists both. kind is direct (the writer and one other person) or group (a group chat; participant.name is the chat's title and memberCount counts everyone). Pass kind=direct to leave group chats out. Each item names the writer's account it belongs to, the other person, a preview of the last message and unreadCount. messagesComplete false means some messages are still being read. Use get_conversation with an id to read the messages. Requires the Revenue Writer plan or above.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "list_conversations"
  }
}

Input Schema

Name	Required	Description	Default
`kind`	No	direct: one-to-one DMs. group: group chats (Substack). all: both.	all
`limit`	No
`state`	No		inbox
`cursor`	No
`platform`	No		all

Output Schema

Name	Required	Description	Default
`page`	Yes
`sync`	Yes
`conversations`	Yes
`connectedAccount`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed4 schema fields changedNov 12, 2026

addedInput schema / properties / kind

Added value: +{
+  "default": "all",
+  "description": "direct: one-to-one DMs. group: group chats (Substack). all: both.",
+  "enum": [
+    "all",
+    "direct",
+    "group"
+  ],
+  "type": "string"
+}

addedOutput schema / properties / conversations / items / properties / kind

Added value: +{
+  "enum": [
+    "direct",
+    "group"
+  ],
+  "type": "string"
+}

addedOutput schema / properties / conversations / items / properties / memberCount

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ]
+}

changedOutput schema / properties / conversations / items / required

Previous value: -[
-  "id",
-  "platform",
-  "account",
-  "participant",
-  "lastMessagePreview",
-  "lastMessageAt",
-  "unreadCount",
-  "state",
-  "url",
-  "messagesComplete"
-]New value: +[
+  "id",
+  "platform",
+  "kind",
+  "memberCount",
+  "account",
+  "participant",
+  "lastMessagePreview",
+  "lastMessageAt",
+  "unreadCount",
+  "state",
+  "url",
+  "messagesComplete"
+]

AddedNov 10, 2026

TDQS

A4.7/5.0

Behavior5/5

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

Although annotations already declare readOnlyHint=true and destructiveHint=false, the description adds material behavior the annotations cannot express: data is cached rather than fetched live, refresh cadence is a few minutes (Substack about hourly), freshness is exposed via sync.lastSyncedAt, and messagesComplete=false signals still-loading messages. The plan-gating requirement (Revenue Writer or above) is a real authorization constraint the agent would otherwise not know.

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

Conciseness4/5

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

Front-loaded correctly with the verb, resource, and ordering, and nearly every sentence adds distinct information with no filler. It is a dense single block of roughly eight sentences, though, and pagination-free structuring with no separation between scoping, semantics, and connectivity would help scanning.

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

Completeness4/5

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

Rich relative to complexity: an output schema exists, so return fields need only light mention, and the description adds sync freshness, completeness signaling, grouping semantics, and plan requirements. The remaining gap is cursor/limit behavior — an agent is not told how to page through results beyond 50 items.

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

Parameters4/5

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

Schema coverage is only 20%, so the description must carry semantics, and it does for three of five parameters: state (with the 7-day inbox rule), kind (direct vs group, explaining that participant.name is the chat title and memberCount counts everyone), and platform via the enumerated source list. limit and cursor are never mentioned, leaving pagination semantics unexplained despite cursor's maxLength constraint in the schema.

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?

Opens with a specific verb and resource ("Read the writer's stored direct-message conversations") and scopes the source platforms by name. It also distinguishes itself from the sibling read path by ending with "Use get_conversation with an id to read the messages," so an agent can separate list-vs-fetch without opening a schema.

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?

Explicit routing rules: state=inbox is the default and covers conversations where the reader wrote last within 7 days, history is "the rest," all is both, and "Pass kind=direct to leave group chats out." The alternative tool is named with the exact condition (an id) that selects it.

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

All Narrareach tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?