Skip to main content
Glama

Uwear

by ai.uwear
OverviewDiscussionsAdmin

Create Brief for Approval

propose_brief
Destructive

Create a Uwear BriefProposal from canonical generation commands. Every commands[].input is the MCP-safe GenerationIntent fields; commands and immutable plans are persisted without translation. Supply a concrete model_slug for generate, edit, upscale, and video. Use durable command.source IDs for uploaded files or generation results, and reference_attachments for additional references. For video, attach available full back or side garment assets that the camera may reveal when capacity permits; having the asset uploaded is not enough. If the response contains video_garment_view_not_attached, explain its exact assets, node, and capacity, then follow its remediation. Never mix reference_attachments with img_ref_urls or append recommendations beyond remaining capacity. Set execute_immediately=true only when the user explicitly asks to run now. Include creative_context for photoshoots and explain the art direction after proposing. For changes to a visible brief, call update_brief with the complete replacement command list. Webhook callback configuration is API-only.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
commandsYesCanonical generation commands. Each command contains the MCP-safe GenerationIntent fields and a durable source reference. Write each input.prompt as a compact positive scene description (setting, lighting, mood, camera, pose — for video, motion and camera movement); without art_direction_id the prompt is the entire creative direction, so give it the full scene. The attached garment/avatar/reference images carry product appearance, so never pad prompts with restated garment construction details, product-fidelity warnings, or negative 'do not' instruction blocks; styling intent (fit, tuck, drape) is fine. For video commands, attach available full back or side garment assets that the camera may reveal through input.reference_attachments when capacity permits. Do not combine reference_attachments with img_ref_urls; uploaded garment assets are not attached automatically. Webhook callback configuration is API-only.
creative_contextNoRequired for photoshoot proposals and brief rewrites. Summarize the shoot-level creative approach that guided garment combos, avatars, prompts, and pipeline steps. By default prefer one cohesive art direction across the photoshoot, e.g. 'urban summer editorial', 'standard grey e-commerce', or 'sporty studio catalog'. Per-look prompts may differ for garment details, pose, framing, or avatar, but should feel part of the same shoot unless the user explicitly asks for multiple art directions, split concepts, A/B routes, or varied campaign directions. The assistant should explain this art direction to the user after proposing the brief and ask if they want changes.
execute_immediatelyNoSet true only when the user explicitly asks to prepare/create/generate/run the photoshoot without another review step. When true, Uwear persists the brief, confirms it, and executes it directly if credits and validation allow; otherwise it falls back to the editable BriefProposal.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • addedInput schema / $defs / GenerationReferenceAttachmentInput / properties / outfit_id
      Added value: +{
+  "default": null,
+  "description": "Outfit reference. The generation uses the front image of every garment in the outfit, as if each were attached as clothing_item.",
+  "minimum": 1,
+  "type": "integer"
+}
    • changedInput schema / $defs / GenerationReferenceAttachmentInput / properties / type / enum
      Previous value: -[
-  "uploaded_file",
-  "generation_result",
-  "location",
-  "avatar",
-  "clothing_item"
-]New value: +[
+  "uploaded_file",
+  "generation_result",
+  "location",
+  "avatar",
+  "clothing_item",
+  "outfit"
+]
  2. Changed1 schema field changed
    • addedInput schema / $defs / GenerationIntent / properties / generate_audio / description
      Added value: +"Include audio in the delivered video. False delivers a file with no audio stream; fixed-audio providers are muted after generation, without reducing native-generation credits."
  3. Changed2 schema fields changed
    • addedInput schema / $defs / GenerationIntent / properties / audio_ref_urls
      Added value: +{
+  "default": null,
+  "description": "Public audio URLs supplied as model references. Only models that declare a positive max_reference_audio capability accept them.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
    • addedInput schema / $defs / GenerationIntent / properties / video_ref_urls
      Added value: +{
+  "default": null,
+  "description": "Public video URLs supplied as model references. Only models that declare a positive max_reference_videos capability accept them.",
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}
  4. Changed1 schema field changed
    • changedInput schema / $defs / GenerationIntent / properties / prompt / description
      Previous value: -"Scene brief for this shot: setting, lighting, mood, camera/framing, pose — 2-5 compact positive sentences. Without art_direction_id this prompt reaches the image model as-is and is the entire creative direction: carry the shoot's world, lighting, and atmosphere here. With art_direction_id, the saved direction carries reusable style; add only per-shot intent. Supplied garment/avatar/reference images carry product appearance: refer to them semantically ('the supplied jacket'), never by image number — the backend appends the numbered reference map at dispatch. Styling intent (fit, tuck, drape, opening state) belongs here when it matters; re-described garment construction, product-fidelity warnings, and negative 'do not ...' blocks do not — with good references they degrade output. Say what to show, not what to avoid. Video commands: describe motion and camera movement. May be empty for upscale; generation routes validate separately."New value: +"Scene brief for this shot: setting, lighting, mood, camera/framing, pose — 2-5 compact positive sentences. Without art_direction_id this prompt reaches the image model as-is and is the entire creative direction: carry the shoot's world, lighting, and atmosphere here. With art_direction_id, the saved direction carries reusable style; add only per-shot intent. Supplied garment/avatar/reference images carry product appearance: refer to them semantically ('the supplied jacket'), never by image number — the backend appends the numbered reference map at dispatch. Styling intent (fit, tuck, drape, opening state) belongs here when it matters; re-described garment construction, product-fidelity warnings, and negative 'do not ...' blocks do not — with good references they degrade output. Say what to show, not what to avoid. Video commands: describe motion and camera movement. Video does not inherit the source generation's ArtDirection; omit art_direction_id to use only the source frame and this prompt. May be empty for upscale; generation routes validate separately."
  5. Changed1 schema field changed
    • addedInput schema / $defs / GenerationIntent / properties / last_frame_attachment
      Added value: +{
+  "$ref": "#/$defs/GenerationReferenceAttachmentInput",
+  "default": null,
+  "description": "Durable reference to the photo the clip closes on, resolved to a URL at dispatch. The durable form of last_frame_url, for callers that hold an id rather than a link — a workflow wiring a step's output into a later step's last frame cannot know the URL when the graph is authored."
+}
  6. Changed5 schema fields changed
    • changedInput schema / $defs / GenerationIntent / description
      Previous value: -"Caller-owned model-generation intent shared by every public adapter.\n\nAuthentication-derived identity deliberately does not belong here. REST,\nAgent, MCP, batch, and integrations all submit this exact contract; the\ncommand planner combines it with server-owned execution context later."New value: +"Caller-owned model-generation intent shared by every public adapter.\n\nAuthentication-derived identity deliberately does not belong here. REST,\nAgent, MCP, batch, and integrations all submit this exact contract; the\ncommand planner combines it with server-owned execution context later. MCP omits webhook callback configuration, which remains API-only."
    • removedInput schema / $defs / GenerationIntent / properties / webhook_events
      Removed value: -{
-  "default": null,
-  "items": {
-    "enum": [
-      "generation.completed",
-      "generation.failed"
-    ],
-    "type": "string"
-  },
-  "type": "array"
-}
    • removedInput schema / $defs / GenerationIntent / properties / webhook_secret
      Removed value: -{
-  "default": null,
-  "description": "Optional HMAC signing secret for customer callbacks.",
-  "type": "string",
-  "ui": {
-    "sensitive": true
-  }
-}
    • removedInput schema / $defs / GenerationIntent / properties / webhook_url
      Removed value: -{
-  "default": null,
-  "type": "string"
-}
    • changedInput schema / properties / commands / description
      Previous value: -"Canonical generation commands. Each command contains the exact REST GenerationIntent contract and a durable source reference. Write each input.prompt as a compact positive scene description (setting, lighting, mood, camera, pose — for video, motion and camera movement); without art_direction_id the prompt is the entire creative direction, so give it the full scene. The attached garment/avatar/reference images carry product appearance, so never pad prompts with restated garment construction details, product-fidelity warnings, or negative 'do not' instruction blocks; styling intent (fit, tuck, drape) is fine. For video commands, attach available full back or side garment assets that the camera may reveal through input.reference_attachments when capacity permits. Do not combine reference_attachments with img_ref_urls; uploaded garment assets are not attached automatically."New value: +"Canonical generation commands. Each command contains the MCP-safe GenerationIntent fields and a durable source reference. Write each input.prompt as a compact positive scene description (setting, lighting, mood, camera, pose — for video, motion and camera movement); without art_direction_id the prompt is the entire creative direction, so give it the full scene. The attached garment/avatar/reference images carry product appearance, so never pad prompts with restated garment construction details, product-fidelity warnings, or negative 'do not' instruction blocks; styling intent (fit, tuck, drape) is fine. For video commands, attach available full back or side garment assets that the camera may reveal through input.reference_attachments when capacity permits. Do not combine reference_attachments with img_ref_urls; uploaded garment assets are not attached automatically. Webhook callback configuration is API-only."
  7. Changed5 schema fields changed
    • removedInput schema / $defs / GenerationCommand / properties / source / discriminator / mapping
      Removed value: -{
-  "generation_result": "#/$defs/GenerationResultCommandSource",
-  "pipeline_step": "#/$defs/PipelineStepCommandSource",
-  "public_url": "#/$defs/PublicUrlCommandSource",
-  "uploaded_file": "#/$defs/UploadedFileCommandSource"
-}
    • removedInput schema / $defs / GenerationResultCommandSource
      Removed value: -{
-  "properties": {
-    "generation_result_id": {
-      "minimum": 1,
-      "type": "integer"
-    },
-    "type": {
-      "const": "generation_result",
-      "type": "string"
-    }
-  },
-  "required": [
-    "type",
-    "generation_result_id"
-  ],
-  "type": "object"
-}
    • removedInput schema / $defs / PipelineStepCommandSource
      Removed value: -{
-  "description": "Durable identity for a source produced by an earlier pipeline step.",
-  "properties": {
-    "output_index": {
-      "minimum": 0,
-      "type": "integer"
-    },
-    "post_claim_reference": {
-      "minLength": 1,
-      "type": "string"
-    },
-    "step_index": {
-      "minimum": 0,
-      "type": "integer"
-    },
-    "type": {
-      "const": "pipeline_step",
-      "type": "string"
-    }
-  },
-  "required": [
-    "type",
-    "post_claim_reference",
-    "step_index",
-    "output_index"
-  ],
-  "type": "object"
-}
    • removedInput schema / $defs / PublicUrlCommandSource
      Removed value: -{
-  "properties": {
-    "image_url": {
-      "minLength": 1,
-      "type": "string"
-    },
-    "type": {
-      "const": "public_url",
-      "type": "string"
-    }
-  },
-  "required": [
-    "type",
-    "image_url"
-  ],
-  "type": "object"
-}
    • removedInput schema / $defs / UploadedFileCommandSource
      Removed value: -{
-  "properties": {
-    "type": {
-      "const": "uploaded_file",
-      "type": "string"
-    },
-    "uploaded_file_id": {
-      "minimum": 1,
-      "type": "integer"
-    }
-  },
-  "required": [
-    "type",
-    "uploaded_file_id"
-  ],
-  "type": "object"
-}
  8. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Beyond the annotations (destructiveHint=true, readOnlyHint=false), the description discloses persistence behavior ('commands and immutable plans are persisted without translation'), error-remediation behavior (video_garment_view_not_attached response handling), and the API-only limitation of webhook configuration. It also describes the execute_immediately fallback behavior via the schema. No contradiction with annotations exists.

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?

The description is dense and long, but every sentence earns its place: core purpose first, then command construction rules, video specifics, error handling, exclusions, and routing to update_brief. It is front-loaded with the primary purpose. Slightly over-long, but well-structured and free of redundant filler.

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 complex tool with nested command schemas and no output schema, the description is remarkably complete. It covers command construction, required fields, source ID semantics, video asset attachment rules, error-response handling, execute_immediately semantics, creative_context requirements, and routing to update_brief. The only missing piece—exact return value—is mitigated by the description's error-handling guidance and the absence of an output schema.

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 the baseline is 3. The description adds meaningful guidance on top: requiring a concrete model_slug for specific command types, mandating durable command.source IDs, clarifying the role of reference_attachments for video, and explaining when execute_immediately and creative_context should be set. This elevates the semantics beyond the schema's plain descriptions.

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?

The description opens with a specific verb and resource: 'Create a Uwear BriefProposal from canonical generation commands.' It clearly distinguishes itself from sibling update_brief by stating 'For changes to a visible brief, call update_brief with the complete replacement command list.' The purpose is unambiguous and differentiated.

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?

The description provides explicit when-to-use guidance: when to set execute_immediately ('only when the user explicitly asks to run now'), when creative_context is required, and it names the alternative tool (update_brief) with the condition selecting it. It also gives negative constraints such as 'Never mix reference_attachments with img_ref_urls' and 'do not append recommendations beyond remaining capacity.'

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

All Uwear toolsInspect
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