Sceneplane

Prepare downloadable texture maps

texture_maps_create

Convert uploaded PNG/JPEG images or supplied maps into a bounded PBR map package without changing the scene. Supplied maps win; derived relief and roughness are labeled approximations. Returns a durable job; job_get.result.textureMaps contains map, ZIP, repeat-preview and manifest artifact IDs. Use image_prompt_assist for source-image guidance.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`roles`	No
`options`	No
`profile`	No		surface
`projectId`	Yes
`sourceMaps`	No
`materialName`	No		Image Material
`sourceAssetId`	No
`idempotencyKey`	Yes	Caller-chosen key. Replaying it with IDENTICAL input returns the EXISTING job instead of queueing a second one — a restarted driver adopts its own in-flight work rather than duplicating or orphaning it. Vary the key to force a new job. Full semantics: the same key with DIFFERENT input is rejected with IDEMPOTENCY_CONFLICT (never silently replayed, so a stale result cannot be mistaken for a fresh one). A key is recorded only once a job is actually created, so a request rejected beforehand (validation, APPROVAL_REQUIRED, quota) leaves the key unused and free to reuse. A key bound to a job keeps resolving to that job even after it fails or is cancelled — use a NEW key to genuinely retry cancelled work. Keys live 48 hours and are scoped per principal+deployment+operation, which means they SPAN SESSIONS: a plain descriptive key like "hero-final-1024-v1" will collide with your own run from yesterday and be rejected. Include a project id or a timestamp. This scope used to be stated only in the conflict error, i.e. after the round trip it cost — a field build lost two that way, ~34 hours after unrelated work had taken the same names.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`job`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed4 schema fields changedSep 27, 2026
- addedInput schema / properties / options / properties / roughnessRange / items
```
Added value: +{
+  "maximum": 1,
+  "minimum": 0,
+  "type": "number"
+}
```
- addedInput schema / properties / options / properties / roughnessRange / maxItems
```
Added value: +2
```
- addedInput schema / properties / options / properties / roughnessRange / minItems
```
Added value: +2
```
- removedInput schema / properties / options / properties / roughnessRange / prefixItems
```
Removed value: -[
-  {
-    "maximum": 1,
-    "minimum": 0,
-    "type": "number"
-  },
-  {
-    "maximum": 1,
-    "minimum": 0,
-    "type": "number"
-  }
-]
```
AddedSep 27, 2026

TDQS

B3.4/5.0

Behavior4/5

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

Beyond the annotations (readOnly=false, destructive=false, openWorld=false), it discloses real behavior: supplied maps take precedence over derived ones, derived relief/roughness are labeled approximations, and the call returns a durable job rather than an inline result. Since annotations already carry the safety profile, this added operational context is meaningful.

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?

Four tight sentences, front-loaded with the core transformation and precedence rule before the return contract and the sibling pointer. Dense domain vocabulary but no filler.

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?

The return contract is covered well — it names job_get.result.textureMaps and the artifact types (map, ZIP, repeat-preview, manifest) — and an output schema exists so return values need not be restated. The gap is on the input side: with 13% schema coverage over a large nested options surface, the description is not complete enough for the agent to drive options confidently.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

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

Schema description coverage is only 13% and the parameters are numerous and complex (a 20+ key options object, roles, profiles, sourceMaps with channel/colorSpace/invert). The description gestures at 'uploaded PNG/JPEG images or supplied maps' and mentions relief/roughness, but leaves roles, profile, sourceAssetId, materialName and nearly all options undocumented in both schema and prose.

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: 'Convert uploaded PNG/JPEG images or supplied maps into a bounded PBR map package without changing the scene.' An agent can tell it produces downloadable texture-map artifacts rather than applying materials, but it never names the closest siblings (material_from_image, material_from_maps) so the boundary is inferred, not explicit.

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?

The description scopes the operation ('without changing the scene') and defers one related need to a sibling ('Use image_prompt_assist for source-image guidance'), which implies when this tool is the right entry point. However, it gives no explicit when-to-use vs material_from_image / material_from_maps, and no stated prerequisites or exclusions.

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

All Sceneplane 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?