Spytrend

Find creatives that look the same

find_similar_creatives

Read-only

Find creatives that look the same as one creative, using the production similarity index. This is visual similarity, not exact creative membership. threshold_status distinguishes an explicitly requested threshold from the production default; result_status=bounded_native_cap means the upstream similarity endpoint exposes at most 20 candidates and has_more is conservative at that cap. Returns metadata without media URLs; use get_media for selected ids. QUOTA: 1 token per delivered row.

Input Schema

TableJSON Schema

Name	Required	Description
`limit`	No	maximum metadata rows to return (default 20, maximum 20)
`threshold`	No	optional maximum visual difference between two creatives (0 uses the production default; maximum 128 — lower is stricter)
`creative_id`	Yes	source creative UUID from search_creatives

Output Schema

TableJSON Schema

Name	Required	Description
`data`	No
`error`	No	Present only when isError is true: machine-readable failure. The human explanation stays in content.
`params`	No	Request parameter echo: applied = parameters that shaped this result; normalized = parameters rewritten before applying (alias, type coercion, resolved id); ignored = parameters that were accepted but NOT applied, with the reason.
`has_more`	No
`threshold`	No
`result_status`	No
`similarity_basis`	No
`threshold_status`	No
`source_creative_id`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedSep 25, 2026

addedOutput schema / properties / error

Added value: +{
+  "description": "Present only when isError is true: machine-readable failure. The human explanation stays in content.",
+  "properties": {
+    "code": {
+      "description": "fine-grained, stable failure code (e.g. invalid_arguments, backend_busy)",
+      "type": "string"
+    },
+    "kind": {
+      "description": "failure class: invalid_argument, not_found, permission_denied, unauthenticated, quota_exceeded, rate_limited, temporarily_unavailable, unavailable_until_ready, unsupported, internal",
+      "type": "string"
+    },
+    "message": {
+      "description": "the same human text as content[0]",
+      "type": "string"
+    },
+    "outcome": {
+      "description": "activity-feed outcome class",
+      "type": "string"
+    },
+    "param": {
+      "description": "the request parameter the failure is about, when known",
+      "type": "string"
+    },
+    "retry_after_seconds": {
+      "description": "wait this long before retrying",
+      "type": "integer"
+    },
+    "retryable": {
+      "description": "true when repeating the SAME call can succeed (after retry_after_seconds when present)",
+      "type": "boolean"
+    }
+  },
+  "type": "object"
+}

addedOutput schema / properties / params

Added value: +{
+  "description": "Request parameter echo: applied = parameters that shaped this result; normalized = parameters rewritten before applying (alias, type coercion, resolved id); ignored = parameters that were accepted but NOT applied, with the reason.",
+  "properties": {
+    "applied": {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    "ignored": {
+      "items": {
+        "properties": {
+          "param": {
+            "type": "string"
+          },
+          "reason": {
+            "type": "string"
+          },
+          "to": {
+            "description": "the parameter it was applied as, when renamed",
+            "type": "string"
+          },
+          "value": {
+            "description": "the value actually applied, when rewritten",
+            "type": "string"
+          }
+        },
+        "type": "object"
+      },
+      "type": "array"
+    },
+    "normalized": {
+      "items": {
+        "properties": {
+          "param": {
+            "type": "string"
+          },
+          "reason": {
+            "type": "string"
+          },
+          "to": {
+            "description": "the parameter it was applied as, when renamed",
+            "type": "string"
+          },
+          "value": {
+            "description": "the value actually applied, when rewritten",
+            "type": "string"
+          }
+        },
+        "type": "object"
+      },
+      "type": "array"
+    }
+  },
+  "type": "object"
+}

removedOutput schema / required

Removed value: -[
-  "source_creative_id",
-  "similarity_basis",
-  "threshold",
-  "threshold_status",
-  "data",
-  "has_more",
-  "result_status"
-]

First observedSep 2, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations declare readOnlyHint=true and destructiveHint=false, so the safety profile is already known. The description adds valuable behavioral context beyond annotations: the upstream cap of 20 candidates, the conservative has_more semantics at that cap, the meaning of threshold_status, the distinction between production default and explicit threshold, and the 1-token-per-row quota. This is rich behavioral disclosure that informs an agent about cost and edge cases.

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?

Description is compact at about 80 words and front-loads the core purpose, then adds behavioral detail. Every sentence earns its place: the first states purpose, the second clarifies the distinction from exact membership, the third explains status values and the cap, the fourth routes to get_media, and the last states quota. Slightly dense, but not wasteful.

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?

The output schema exists, so return values need not be described in prose. The description covers purpose, edge cases (cap, conservative has_more), quota cost, and follow-up tooling. For a read-only similarity search tool with a small parameter surface, 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.

Parameters4/5

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

Schema description coverage is 100%, so all three parameters are documented in the schema itself, which is a baseline-3 situation. The description adds extra meaning beyond the schema by clarifying the semantics of threshold (visual difference, lower is stricter) and explaining why limit is capped at 20 (upstream endpoint cap), which helps the agent reason about parameter interactions.

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?

Description states a specific verb+resource: find creatives that look the same as one creative, using the production similarity index. It distinguishes from exact creative membership and from the sibling find_similar_ads and find_similar_webmasters by specifying creatives and the visual similarity basis.

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?

Explicitly differentiates visual similarity from exact creative membership, which tells the agent when this tool is appropriate vs search_creatives. It also names the threshold_status and result_status states to clarify behavior. It directs to get_media for media URLs, routing the agent to the correct follow-up tool.

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

All Spytrend 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.