campaignstack

campaignstack_send_message

Destructive

Ask CampaignStack to send one message into a LinkedIn conversation the owner's connected account is already part of, optionally with attachments given as public URLs. Takes a conversationRef { platform, id } from campaignstack_get_conversations. Read the thread with campaignstack_get_conversations using ids and include.messages first. If the last message is the owner's and the other person has not answered since, do not send: say so, and send only if the user confirms they want to follow up again. This reaches a real person. Before calling it, tell the user exactly who it goes to and read the wording back to them, and wait for an explicit yes. The send is dispatched asynchronously and the response carries a messageEntityUrn; campaignstack_get_conversations reports delivery. Pass sendAt to hold the message and send it later instead of now; it is cancelled if the other person replies first, so read the wording and the time back and wait for a yes. campaignstack_get_conversations with include.scheduledMessages lists pending ones; campaignstack_edit_conversations cancels or edits one. One person per call. There is no bulk form, and calling this in a loop over a list is not the supported way to reach a list: add the leads with campaignstack_edit_lead_list and run a workflow over it with campaignstack_trigger_workflow, which paces the sends and applies the review step the workspace configured. Sending is paced by CampaignStack's own daily and weekly limits and business hours, which are set by a human in the app and are not writable from here; see overrideOwnLimits before asking for an exception.

Input Schema

TableJSON Schema

Name	Required	Description
`text`	Yes
`sendAt`	No	ISO 8601 with a UTC offset. Omit to send now. With it, the message is held and sent at that time instead (1 minute to 180 days out) - cancelled automatically if the other person replies first. Read the wording and the time back to the user and wait for a yes before calling this with sendAt set.
`attachments`	No	Up to 5 files to attach (images or documents). Downloaded server-side and validated against the LinkedIn attachment policy.
`workspaceId`	No	Workspace ID. Omit it when this connection reaches one workspace: a workspace key (cs_) is bound to its own, and a user key (csu_) or connected app whose person belongs to exactly one workspace defaults to that one. Required only when the person belongs to several; the error then lists them, and campaignstack_whoami reports them too.
`conversationRef`	Yes
`overrideOwnLimits`	No	Tells CampaignStack to skip its OWN protective pacing for this one call: the daily budget, the weekly cap, the business-hours window, the account status check and the risk pause. These are limits CampaignStack sets to keep the account healthy. They are not LinkedIn's controls, and nothing here touches, negotiates with or works around anything LinkedIn enforces. LinkedIn's own measured ceiling is a separate check that is never overridable, and neither is a pause the account's owner set. Default false, and it should stay false. Setting it true voids the safety guarantee: from that call on, CampaignStack can no longer promise the pacing that keeps the account out of trouble, and the risk of a restriction sits with the owner. Every use is recorded in the workspace audit log. Ask the owner in plain words and get an explicit yes first. A refusal from the safety system is that system working, not a fault to route around: report the reason and let the person decide, and never retry the same send with this set.
`dangerouslyBypassSafetySystem`	No	Deprecated name for overrideOwnLimits. Use overrideOwnLimits.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`sendAt`	No
`status`	No
`messageEntityUrn`	No
`scheduledMessageId`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedOct 6, 2026

addedInput schema / properties / sendAt

Added value: +{
+  "description": "ISO 8601 with a UTC offset. Omit to send now. With it, the message is held and sent at that time instead (1 minute to 180 days out) - cancelled automatically if the other person replies first. Read the wording and the time back to the user and wait for a yes before calling this with sendAt set.",
+  "format": "date-time",
+  "type": "string"
+}

addedOutput schema / properties / scheduledMessageId

Added value: +{
+  "anyOf": [
+    {
+      "anyOf": [
+        {
+          "not": {}
+        },
+        {
+          "type": "string"
+        }
+      ]
+    },
+    {
+      "type": "null"
+    }
+  ]
+}

addedOutput schema / properties / sendAt

Added value: +{
+  "anyOf": [
+    {
+      "anyOf": [
+        {
+          "not": {}
+        },
+        {
+          "type": "string"
+        }
+      ]
+    },
+    {
+      "type": "null"
+    }
+  ]
+}

Changed1 schema field changedSep 29, 2026

addedInput schema / properties / workspaceId

Added value: +{
+  "description": "Workspace ID. Omit it when this connection reaches one workspace: a workspace key (cs_) is bound to its own, and a user key (csu_) or connected app whose person belongs to exactly one workspace defaults to that one. Required only when the person belongs to several; the error then lists them, and campaignstack_whoami reports them too.",
+  "type": "string"
+}

Changed1 schema field changedSep 20, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": true,
+  "properties": {
+    "messageEntityUrn": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "type": "string"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "status": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "type": "string"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    }
+  },
+  "type": "object"
+}

Changed3 schema fields changedSep 14, 2026

removedInput schema / properties / dangerouslyBypassSafetySystem / default
```
Removed value: -false
```

addedInput schema / properties / dangerouslyBypassSafetySystem / description

Added value: +"Deprecated name for overrideOwnLimits. Use overrideOwnLimits."

addedInput schema / properties / overrideOwnLimits

Added value: +{
+  "description": "Tells CampaignStack to skip its OWN protective pacing for this one call: the daily budget, the weekly cap, the business-hours window, the account status check and the risk pause. These are limits CampaignStack sets to keep the account healthy. They are not LinkedIn's controls, and nothing here touches, negotiates with or works around anything LinkedIn enforces. LinkedIn's own measured ceiling is a separate check that is never overridable, and neither is a pause the account's owner set. Default false, and it should stay false. Setting it true voids the safety guarantee: from that call on, CampaignStack can no longer promise the pacing that keeps the account out of trouble, and the risk of a restriction sits with the owner. Every use is recorded in the workspace audit log. Ask the owner in plain words and get an explicit yes first. A refusal from the safety system is that system working, not a fault to route around: report the reason and let the person decide, and never retry the same send with this set.",
+  "type": "boolean"
+}

First observedSep 4, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations only declare destructiveHint/openWorldHint/non-idempotent; the description adds far more: asynchronous dispatch with a messageEntityUrn response, delivery reported by get_conversations, sendAt holding with auto-cancel on reply, workspace-owned pacing limits that are not writable from here, and the overrideOwnLimits caveat that overrides are audited and void the safety guarantee. This is exactly the extra behavioral context annotations cannot carry.

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 with the core purpose, then prerequisites, then the safety/confirmation rules, then the bulk alternative and pacing. Every section is substantive, but the 'read the wording back and wait for a yes' instruction is repeated for both the immediate and sendAt cases, and the piece runs long, which costs a point.

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?

With an output schema present the description need not explain return values, and it still mentions the messageEntityUrn. Combined with prerequisites, confirmation requirements, scheduling semantics, pacing limits and the bulk-workflow alternative, an agent has everything needed to call this a 7-parameter mutation correctly.

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 71%, with rich inline descriptions for sendAt, attachments, workspaceId and overrideOwnLimits, so baseline is around 3. The description adds meaning beyond the schema by naming the source of conversationRef (campaignstack_get_conversations) and the 'one person per call' constraint. It adds little for text, but on balance it pushes past the baseline.

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?

States a specific verb and resource with scope: 'send one message into a LinkedIn conversation the owner's connected account is already part of', plus the attachment capability. It is clearly distinguishable from siblings like campaignstack_send_inmail, campaignstack_send_connection_request and campaignstack_craft_message without opening any 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?

Gives explicit prerequisites (read the thread with campaignstack_get_conversations using ids and include.messages first), an explicit when-not (do not send if the last message is the owner's with no reply), the confirmation ritual before calling, and named alternatives for the bulk case (campaignstack_edit_lead_list plus campaignstack_trigger_workflow). Nothing is left to inference.

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

All campaignstack 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?