trident-mcp

Overview Schema Source Related Servers Discussions Admin

text_to_image

Destructive

Generate a reference image from a text prompt. Returns a task ID for immediate use as image_to_model input, enabling text-to-3D workflows.

Instructions

Generate a reference image from a text prompt. Its task ID can be passed directly as image_to_model input. Creates a Tripo task (spends credits) and returns its task_id immediately; then call wait_for_task.

Input Schema

TableJSON Schema

Name	Required	Description
`size`	No	Output size: 2K, 4K or WxH such as 2048x2048 (default)
`model`	No	Image model: seedream_v4 (text_to_image default), seedream_v5 (image_to_image default), banana, banana2, banana_pro, chat_image_2, chat_image_2.5_flare, chat_image_2.5_sunburst. See list_models
`prompt`	Yes	What to draw. For 3D references: one object, plain background. Append --no <things> to exclude things
`quality`	No	chat_image models only (default low); xhigh and max need a 2.5 model
`template`	No	Preset prompt template, e.g. t_pose for rig-ready characters
`watermark`	No	Add a watermark (seedream only)
`background`	No	chat_image 2.5 models only; transparent needs output_format png
`aspect_ratio`	No	banana models only (default 1:1); 1:8, 1:4, 4:1 and 8:1 need banana2
`output_format`	No	Image file format (default png)

Output Schema

TableJSON Schema

Name	Required	Description	Default
`model`	No
`task_id`	Yes
`produces`	Yes
`warnings`	No
`operation`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed22 schema fields changedv0.4.0Oct 19, 2026

addedInput schema / properties / aspect_ratio

Added value: +{
+  "description": "banana models only (default 1:1); 1:8, 1:4, 4:1 and 8:1 need banana2",
+  "enum": [
+    "1:1",
+    "2:3",
+    "3:2",
+    "3:4",
+    "4:3",
+    "4:5",
+    "5:4",
+    "9:16",
+    "16:9",
+    "21:9",
+    "1:8",
+    "1:4",
+    "4:1",
+    "8:1"
+  ],
+  "type": "string"
+}

addedInput schema / properties / background

Added value: +{
+  "description": "chat_image 2.5 models only; transparent needs output_format png",
+  "enum": [
+    "auto",
+    "opaque",
+    "transparent"
+  ],
+  "type": "string"
+}

removedInput schema / properties / height

Removed value: -{
-  "description": "Optional output height",
-  "type": "integer"
-}

changedInput schema / properties / model / description

Previous value: -"Image model, for example seedream_v5, gemini-3-pro, or chat_image_2"New value: +"Image model: seedream_v4 (text_to_image default), seedream_v5 (image_to_image default), banana, banana2, banana_pro, chat_image_2, chat_image_2.5_flare, chat_image_2.5_sunburst. See list_models"

removedInput schema / properties / negativePrompt

Removed value: -{
-  "description": "What to avoid in the generated image",
-  "type": "string"
-}

addedInput schema / properties / output_format

Added value: +{
+  "description": "Image file format (default png)",
+  "enum": [
+    "png",
+    "jpeg"
+  ],
+  "type": "string"
+}

changedInput schema / properties / prompt / description

Previous value: -"Text prompt for image generation"New value: +"What to draw. For 3D references: one object, plain background. Append --no <things> to exclude things"

addedInput schema / properties / quality

Added value: +{
+  "description": "chat_image models only (default low); xhigh and max need a 2.5 model",
+  "enum": [
+    "low",
+    "medium",
+    "high",
+    "xhigh",
+    "max"
+  ],
+  "type": "string"
+}

removedInput schema / properties / seed

Removed value: -{
-  "description": "Optional deterministic seed",
-  "type": [
-    "null",
-    "integer"
-  ]
-}

changedInput schema / properties / size / description

Previous value: -"Optional output size string if supported by the selected image model"New value: +"Output size: 2K, 4K or WxH such as 2048x2048 (default)"

changedInput schema / properties / template / description

Previous value: -"Optional Tripo image-generation template such as asset_extraction or t_pose"New value: +"Preset prompt template, e.g. t_pose for rig-ready characters"

addedInput schema / properties / template / enum

Added value: +[
+  "asset_extraction",
+  "character_completion",
+  "t_pose",
+  "variants",
+  "figure"
+]

addedInput schema / properties / watermark

Added value: +{
+  "description": "Add a watermark (seedream only)",
+  "type": [
+    "null",
+    "boolean"
+  ]
+}

removedInput schema / properties / width

Removed value: -{
-  "description": "Optional output width",
-  "type": "integer"
-}

addedOutput schema / properties / model
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / operation
```
Added value: +{
+  "type": "string"
+}
```
addedOutput schema / properties / produces
```
Added value: +{
+  "type": "string"
+}
```

removedOutput schema / properties / status

Removed value: -{
-  "type": "string"
-}

removedOutput schema / properties / taskId

Removed value: -{
-  "type": "string"
-}

addedOutput schema / properties / task_id
```
Added value: +{
+  "type": "string"
+}
```

addedOutput schema / properties / warnings

Added value: +{
+  "items": {
+    "type": "string"
+  },
+  "type": [
+    "null",
+    "array"
+  ]
+}

changedOutput schema / required

Previous value: -[
-  "taskId",
-  "status"
-]New value: +[
+  "task_id",
+  "operation",
+  "produces"
+]

Addedv0.3.0Jul 11, 2026

TDQS

A3.8/5.0

Behavior4/5

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

Annotations declare non-read-only, destructive, non-idempotent, open-world, and the description adds genuinely useful context beyond them: the operation spends credits, is asynchronous ('returns its task_id immediately'), and requires a follow-up wait_for_task call. It could go further on whether failures refund credits, but the added behavioral value is real.

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

Conciseness5/5

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

Three tight sentences with zero filler. The core purpose and the credit-spending warning are front-loaded, and the async follow-up is stated last in execution order.

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?

An output schema exists, so return values need no explanation, yet the description still usefully states that a task_id is returned and how to consume it. Combined with the credit and async notes, the definition is complete enough to call correctly, with only the sibling-differentiation gap remaining.

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 all 9 parameters with enums and defaults. The description adds no parameter-level meaning of its own, so the baseline of 3 applies.

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?

States a specific verb and resource ('Generate a reference image from a text prompt') and names a downstream consumer (image_to_model). However, it does not differentiate itself from the sibling image_to_image, which also produces images, so an agent must infer the distinction from the name alone.

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

Usage Guidelines3/5

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

Provides a clear workflow cue ('then call wait_for_task') and notes the task ID can feed image_to_model, which implies usage. But it gives no explicit when-to-use vs. image_to_image or other generation siblings, leaving the selection to inference.

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