campaignstack

campaignstack_explain_draft

Read-only

Why a draft says what it says: each sentence attributed to the piece of configuration that produced it (the offer context, a playbook section, the campaign guidance, an agent's script, the outreach intent, the voice profile, a node's instructions, an agent memory, the contact details, the calendar availability, the conversation so far, the product's built-in rules, the lead's own facts), plus the tool and field that changes each one. Works on in-app and MCP-crafted drafts alike. Every citation is re-checked against the prompt section it names, so an unverifiable one comes back as unattributed rather than as a plausible guess. Call this when a draft reads wrong and you need to know WHICH setting to edit; campaignstack_preview_craft_prompt shows the same prompt for a draft that has not been written yet. Pass include ["attempts"] to see whether a draft held for review was rewritten on each attempt and why each one was rejected. Charges mcp_craft credits per call; campaignstack_estimate_cost gives the number before you commit. One model call over a stored draft and its prompt; writes nothing.

Input Schema

Name	Required	Description
`entryId`	No	A current review entry id from campaignstack_get_reviews. Resolves to the draft's recorded craft session. Use craftAuditLogId for historical drafts.
`include`	No	Extra sections. "attempts": every draft the review loop wrote for this message, in order, with a text hash, word count and each check's verdict and objection.
`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.
`craftAuditLogId`	No	A craft audit log row id, usually copied from an internal audit export or previous explain-draft result.

Output Schema

Name	Required	Description	Default
`draft`	No
`fixes`	No
`caller`	No
`verdict`	No
`attempts`	No
`createdAt`	No
`sentences`	No
`craftAuditLogId`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 20, 2026

changedInput schema / properties / entryId / description

Previous value: -"A workflow entry id, e.g. from campaignstack_get_reviews. Resolves to the newest accepted draft for that entry's lead."New value: +"A current review entry id from campaignstack_get_reviews. Resolves to the draft's recorded craft session. Use craftAuditLogId for historical drafts."

Changed2 schema fields changedOct 14, 2026

addedInput schema / properties / include

Added value: +{
+  "description": "Extra sections. \"attempts\": every draft the review loop wrote for this message, in order, with a text hash, word count and each check's verdict and objection.",
+  "items": {
+    "enum": [
+      "attempts"
+    ],
+    "type": "string"
+  },
+  "maxItems": 25,
+  "type": "array"
+}

addedOutput schema / properties / attempts

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

Changed1 schema field changedSep 29, 2026

changedInput schema / properties / workspaceId / description

Previous value: -"Workspace ID. REQUIRED for a user key (csu_), which spans every workspace you belong to; omitted for a workspace key (cs_), which is bound to one and ignores this argument. campaignstack_whoami reports which kind of key you are holding and, for a user key, the workspaces it can reach."New value: +"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."

Changed2 schema fields changedSep 26, 2026

changedInput schema / properties / craftAuditLogId / description

Previous value: -"A row id from campaignstack_export_craft_audit."New value: +"A craft audit log row id, usually copied from an internal audit export or previous explain-draft result."

changedInput schema / properties / entryId / description

Previous value: -"A workflow entry id, e.g. from campaignstack_list_pending_reviews. Resolves to the newest accepted draft for that entry's lead."New value: +"A workflow entry id, e.g. from campaignstack_get_reviews. Resolves to the newest accepted draft for that entry's lead."

Changed1 schema field changedSep 20, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": true,
+  "properties": {
+    "caller": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "type": "string"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "craftAuditLogId": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "type": "string"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "createdAt": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "type": "number"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "draft": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "type": "string"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "fixes": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "items": {},
+              "type": "array"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "sentences": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "items": {},
+              "type": "array"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "verdict": {
+      "anyOf": [
+        {
+          "anyOf": [
+            {
+              "not": {}
+            },
+            {
+              "type": "string"
+            }
+          ]
+        },
+        {
+          "type": "null"
+        }
+      ]
+    }
+  },
+  "type": "object"
+}

AddedSep 13, 2026

TDQS

A4.6/5.0

Behavior5/5

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

Annotations already cover the safety profile (readOnly, non-destructive, closed-world), and the description adds genuinely new context: it charges mcp_craft credits per call with estimate_cost as the pre-check, makes exactly one model call over a stored draft, and writes nothing. It also discloses a non-obvious behavior – citations are re-verified against the named prompt section, so unverifiable ones return as unattributed rather than a plausible guess.

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-loads the core purpose before the alternatives and cost caveats, and every sentence carries information. The parenthetical list of ~14 attribution sources is long and slightly dense, but it does define the tool's scope, so it largely earns its place.

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 diagnostic tool with an output schema (so return shape needn't be explained), the description covers purpose, trigger, alternative, cost, side-effect profile, and the verification caveat. Nothing an agent needs to call it correctly is missing.

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 description coverage is 100%, so the schema already documents entryId vs craftAuditLogId and the workspaceId fallback logic, and the description's mention of include ['attempts'] largely restates the schema's own description. The description adds only marginal framing beyond the structured fields, so the baseline 3 applies.

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+resource ('explain why a draft says what it says') and enumerates the config sources each sentence is attributed to, plus the tool/field that changes each one. It explicitly distinguishes itself from the sibling campaignstack_preview_craft_prompt (draft not yet written) so an agent can route 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?

Gives an explicit trigger ('Call this when a draft reads wrong and you need to know WHICH setting to edit') and names the alternative plus the condition that selects it ('preview_craft_prompt shows the same prompt for a draft that has not been written yet'). It also tells the agent when to pass include ['attempts'].

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?