Cleotic

Save prompt

save_prompt

Add a prompt to a segment, or with prompt_id edit one. Changing an existing prompt's text replaces it, keeping the old prompt's answers in its history. Cost: an active prompt is plan capacity; new, replaced and resumed prompts get an answer from every model straight away, using model budget.

Input Schema

TableJSON Schema

Name	Required	Description
`tags`	No
`text`	No
`brand`	Yes	Brand or study: its ID from list_brands, or its exact name. A partial name is refused.
`market`	No	A free-text label stored on the prompt. It does not change where the prompt is collected: the segment's geo does.
`status`	No
`persona`	No
`segment`	No	The segment (ID or name) for a new prompt.
`category`	No
`language`	No
`prompt_id`	No	Edit this prompt, from list_prompts. Leave out to add one.
`intent_type`	No
`funnel_stage`	No
`idempotency_key`	Yes	A unique key for this intended action. Reuse it with identical arguments when retrying: the write is not made again, and the reply is the first call's resource as saved (replayed: true) with current status; read tools show the resource's current state. Use a new key for another action.

Output Schema

TableJSON Schema

Name	Required	Description
`data`	Yes
`generated_at`	Yes	RFC 3339 time the response was made.
`schema_version`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed7 schema fields changedOct 1, 2026

changedInput schema / properties / brand / description

Previous value: -"Brand or study: its ID from list_brands, or its name."New value: +"Brand or study: its ID from list_brands, or its exact name. A partial name is refused."

changedInput schema / properties / idempotency_key / description

Previous value: -"A unique key for this intended action. Reuse it with identical arguments when retrying; use a new key for another action."New value: +"A unique key for this intended action. Reuse it with identical arguments when retrying: the write is not made again, and the reply is the first call's resource as saved (replayed: true) with current status; read tools show the resource's current state. Use a new key for another action."

addedInput schema / properties / market / description

Added value: +"A free-text label stored on the prompt. It does not change where the prompt is collected: the segment's geo does."

changedOutput schema / properties / data / properties / brand / description

Previous value: -"The brand written to; null for brand setup before approval."New value: +"The brand written to; null for start_brand_setup, which has no brand yet."

changedOutput schema / properties / data / properties / cost / properties / readiness_scans / description

Previous value: -"Site readiness scans used from the plan's monthly allowance."New value: +"Full site readiness scans started; counted for the month, with no monthly allowance."

addedOutput schema / properties / data / properties / created_at

Added value: +{
+  "description": "Set on a replay: when the original call was made.",
+  "type": "string"
+}

changedOutput schema / properties / data / properties / replayed / description

Previous value: -"True when this repeats an earlier call with the same idempotency_key."New value: +"True when this repeats an earlier call with the same idempotency_key. The write is not made again: resource is as first saved, status, progress and message are current, and the read tools show the resource's current state."

Changed1 schema field changedOct 1, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "additionalProperties": false,
+  "properties": {
+    "data": {
+      "additionalProperties": false,
+      "properties": {
+        "brand": {
+          "additionalProperties": false,
+          "description": "The brand written to; null for brand setup before approval.",
+          "properties": {
+            "id": {
+              "type": "string"
+            },
+            "name": {
+              "type": "string"
+            },
+            "url": {
+              "description": "The brand's website, or empty when none is set.",
+              "type": "string"
+            }
+          },
+          "required": [
+            "id",
+            "name",
+            "url"
+          ],
+          "type": [
+            "null",
+            "object"
+          ]
+        },
+        "cost": {
+          "additionalProperties": false,
+          "description": "What the write used from the plan's capacity, allowances and model budget; null when nothing.",
+          "properties": {
+            "content_generations": {
+              "description": "Content Studio generations used from the plan's allowance.",
+              "type": "integer"
+            },
+            "model_budget": {
+              "description": "Whether the call spends the organisation's AI model budget.",
+              "type": "boolean"
+            },
+            "note": {
+              "type": "string"
+            },
+            "prompt_capacity": {
+              "description": "Active prompts added against the plan's prompt limit.",
+              "type": "integer"
+            },
+            "queued_model_runs": {
+              "description": "Answer collections queued; each uses model budget.",
+              "type": "integer"
+            },
+            "readiness_checks": {
+              "description": "Readiness re-checks used from the plan's monthly allowance.",
+              "type": "integer"
+            },
+            "readiness_scans": {
+              "description": "Site readiness scans used from the plan's monthly allowance.",
+              "type": "integer"
+            }
+          },
+          "required": [
+            "model_budget",
+            "content_generations",
+            "readiness_scans",
+            "readiness_checks",
+            "prompt_capacity",
+            "queued_model_runs",
+            "note"
+          ],
+          "type": [
+            "null",
+            "object"
+          ]
+        },
+        "message": {
+          "description": "What happens next.",
+          "type": "string"
+        },
+        "operation_id": {
+          "description": "Pass to get_operation to follow the write.",
+          "type": "string"
+        },
+        "poll_after_seconds": {
+          "description": "While pending, how long to wait before calling get_operation; 0 otherwise.",
+          "type": "integer"
+        },
+        "progress": {
+          "additionalProperties": false,
+          "description": "Set while the work runs.",
+          "properties": {
+            "done": {
+              "type": "integer"
+            },
+            "message": {
+              "type": "string"
+            },
+            "stage": {
+              "type": "string"
+            },
+            "total": {
+              "description": "0 when the total is not known.",
+              "type": "integer"
+            }
+          },
+          "required": [
+            "stage",
+            "done",
+            "total",
+            "message"
+          ],
+          "type": [
+            "null",
+            "object"
+          ]
+        },
+        "replayed": {
+          "description": "True when this repeats an earlier call with the same idempotency_key.",
+          "type": "boolean"
+        },
+        "resource": {
+          "additionalProperties": false,
+          "properties": {
+            "id": {
+              "type": "string"
+            },
+            "replaces_prompt_id": {
+              "type": "string"
+            },
+            "segment": {
+              "additionalProperties": false,
+              "properties": {
+                "id": {
+                  "type": "string"
+                },
+                "name": {
+                  "type": "string"
+                }
+              },
+              "required": [
+                "id",
+                "name"
+              ],
+              "type": "object"
+            },
+            "status": {
+              "type": "string"
+            },
+            "tags": {
+              "items": {
+                "type": "string"
+              },
+              "type": "array"
+            },
+            "text": {
+              "type": "string"
+            }
+          },
+          "required": [
+            "id",
+            "text",
+            "status",
+            "segment",
+            "tags",
+            "replaces_prompt_id"
+          ],
+          "type": "object"
+        },
+        "status": {
+          "description": "completed: done. pending: saved, with work still running; call get_operation after poll_after_seconds. failed: the work after the write failed; message says why.",
+          "type": "string"
+        }
+      },
+      "required": [
+        "operation_id",
+        "status",
+        "replayed",
+        "brand",
+        "resource",
+        "progress",
+        "poll_after_seconds",
+        "message",
+        "cost"
+      ],
+      "type": "object"
+    },
+    "generated_at": {
+      "description": "RFC 3339 time the response was made.",
+      "type": "string"
+    },
+    "schema_version": {
+      "type": "string"
+    }
+  },
+  "required": [
+    "schema_version",
+    "generated_at",
+    "data"
+  ],
+  "type": "object"
+}

AddedSep 30, 2026

TDQS

A4/5.0

Behavior5/5

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

Annotations only cover the safety flags; the description goes well beyond them by disclosing the replacement semantics ('changing an existing prompt's text replaces it, keeping the old prompt's answers in its history') and the cost model (plan capacity, immediate answers from every model, model budget). These are exactly the side effects an agent needs before mutating state, and they reconcile with destructiveHint=false.

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?

Three sentences, front-loaded with the create/edit distinction before the cost detail. The final cost sentence is a dense run-on packing in three separate behaviors ('new, replaced and resumed prompts get an answer from every model straight away, using model budget'), but nothing is redundant.

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

Completeness3/5

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

For a 13-parameter mutation with an output schema already covering return values, the description handles behavior and cost well but is thin on the bulk of the optional metadata fields. An agent can safely decide whether to call it, but cannot reliably fill tags, persona, category, language, intent_type or funnel_stage from this definition.

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 only 38% (brand, market, segment, prompt_id, idempotency_key), so eight parameters — tags, text, status, persona, category, language, intent_type, funnel_stage — are undocumented in the schema. The description adds meaning only for text (replacement behavior), status (active/paused cost) and the segment target; the remaining five-plus metadata fields are left entirely unexplained, so it only partially compensates.

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

Purpose4/5

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

The description states a specific verb and resource and clearly splits the two modes: 'Add a prompt to a segment, or with prompt_id edit one.' An agent can tell create from edit without opening the schema. It does not, however, name or distinguish itself from siblings such as suggest_prompts or list_prompts, which matters for an agent choosing between generating, listing and saving prompts.

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

Usage Guidelines4/5

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

It gives clear conditional guidance: omit prompt_id to add, supply it to edit, and it notes the cost consequence that selects one path over another ('an active prompt is plan capacity'). What is missing is any when-not guidance or explicit pointer to the alternative tools (suggest_prompts, list_prompts) for the same job.

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

All Cleotic 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?