kinocut

Overview Schema Source Related Servers Discussions Admin

hyperframes_remove_background

Cut a person or product out of a still image or video by removing the background, with separate models for people versus non-person objects.

Instructions

Cut a person or a product out of a still or video.

Default model u2net_human_seg is people-only (talking heads). For products, packaging, jewelry, bottles, ceramics, or any non-person object, pass model="birefnet-general" (requires pip install "kinocut[object-matte]"). Never falls through to the people model. Call info=True with no input to list models and cache without downloading. Guide: docs/PRODUCT_MATTE.md.

Input Schema

Name	Required	Description	Default
`info`	No
`model`	No
`device`	No		auto
`quality`	No		balanced
`overwrite`	No	Explicitly allow replacing existing media. Existing non-media files are protected.
`input_path`	No
`output_path`	No
`mask_interval`	No
`equipment_overlay`	No
`background_output_path`	No
`fail_if_equipment_on_subject`	No

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv1.16.4Oct 24, 2026

addedInput schema / properties / overwrite

Added value: +{
+  "default": false,
+  "description": "Explicitly allow replacing existing media. Existing non-media files are protected.",
+  "title": "Overwrite",
+  "type": "boolean"
+}

Changed9 schema fields changedv1.16.0Oct 3, 2026

addedInput schema / additionalProperties
```
Added value: +false
```

addedInput schema / properties / equipment_overlay

Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "title": "Equipment Overlay"
+}

addedInput schema / properties / fail_if_equipment_on_subject

Added value: +{
+  "default": false,
+  "title": "Fail If Equipment On Subject",
+  "type": "boolean"
+}

addedInput schema / properties / input_path / anyOf

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

addedInput schema / properties / input_path / default
```
Added value: +null
```
removedInput schema / properties / input_path / type
```
Removed value: -"string"
```

addedInput schema / properties / mask_interval

Added value: +{
+  "default": 1,
+  "title": "Mask Interval",
+  "type": "integer"
+}

addedInput schema / properties / model

Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "title": "Model"
+}

removedInput schema / required
```
Removed value: -[
-  "input_path"
-]
```

Addedv1.5.1Jun 4, 2026

TDQS

B3.1/5.0

Behavior3/5

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

No annotations exist, so the description must carry the full burden. It usefully discloses that the default model won't fall through to people, that birefnet-general requires an extra pip extra, and that info=True lists models without downloading. It says nothing about GPU/device behavior, cost, mask_interval effects, or what equipment_overlay does, which for an 11-param tool is a real gap.

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 core capability is front-loaded in one sentence, and the model-selection rule follows immediately. It is reasonably tight, though the heavy use of RST backticks and stray line breaks slightly hurts scanability.

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

Completeness2/5

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 needn't be covered, but for an 11-parameter tool with 9% schema coverage, several cryptic knobs are explained nowhere. The agent would have to guess at mask_interval, equipment_overlay, and fail_if_equipment_on_subject semantics before calling it.

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 9%, so the description should compensate, yet it only explains model and info. Eight other parameters (device, quality, mask_interval, equipment_overlay, background_output_path, fail_if_equipment_on_subject, input_path, output_path) get no meaning from either source; the schema's bare types/defaults don't tell an agent what mask_interval or equipment_overlay actually do.

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 gives a concrete verb and resource: 'Cut a person or a product out of a still or video', which makes the capability obvious. It doesn't contrast with close siblings such as video_chroma_key, video_luma_key, or video_apply_mask, so the agent must infer which matte tool to use.

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?

Within the tool it routes usage well: the default model is described as people-only and birefnet-general is prescribed for products, packaging, jewelry, etc. But it never says when to prefer this tool over the many sibling masking/keying/removal tools, so the selection guidance is only implied at the sibling level.

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

Deploy Server