Name: Sirv AI Studio
Author: sirv

Sirv AI Studio MCP

AI image, video and 3D tools, assets, products and Shopify for Sirv AI Studio.

Grade A Remote or local · Streamable HTTP · stdio Remote: OAuth sign-in · Package: no sign-in 53 tools v1.1.1 MIT

No reviews yet by sirvnpm Runs on your machine or at its publisher’s address — not through us

93/100

trust score

tools declared

local

or remote

—

downloads a day

What it does

AI image, video and 3D tools, assets, products and Shopify for Sirv AI Studio.

Quickstart

# 1 — add it to Claude Code
claude mcp add --scope user sirv-ai-studio-mcp '--' npx -y sirv-ai-mcp-server

# 2 — ask your agent for something these tools do
#   sirv_remove_background, sirv_upscale, sirv_background_replace and 50 more

Remote: OAuth sign-in · Package: no sign-in.

Sirv AI Studio costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.

Publishernot claimed

sirv1 server

1 category

All 1 server

Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.

Installation

Where are you running it?

Every route below installs the same thing and ends at the same approval screen. This one can run either way — at an address its publisher runs, or as a process under your own user, with your files and your network — and this marketplace is in the path of neither.

Claude Codedeclared by this listing

This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.

claude mcp add --scope user sirv-ai-studio-mcp '--' npx -y sirv-ai-mcp-server

Reconnect, or start a new session, and the tools appear in the model’s tool list.

Or connect to its publisher’s address

Its publisher also runs it at https://mcp.sirv.studio: nothing to start on your machine, and this marketplace is not in the path.

claude mcp add --scope user --transport http sirv-ai-studio-mcp https://mcp.sirv.studio

Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.

What it asks of you

Nothing is required, but it can take SIRV_AI_API_KEY (Studio API key. Without it, the server signs in through the browser.) if you want to set it.

What it writes

One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.

What you approve

53 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.

Tool surface

What the model actually sees. Descriptions are diffed on every release — see version history.

Only tools that write or reach the network

Read from its source code. A server that builds its tool list when it starts can list more tools than this.

Parameter	Type	Required	Notes
image_url	string	yes	URL of the image to process
provider	string	no	—
model	string	no	Model to use for background removal. 'General Use (Heavy)' is recommended for best quality
operating_resolution	string	no	Processing resolution. Higher = better quality but slower

Parameter	Type	Required	Notes
image_url	string	yes	URL of the image to upscale
scale	integer	no	Upscale factor: 2x, 3x, or 4x
model	string	no	Upscaler model: 'esrgan' (fast, general), 'clarity' (AI-enhanced), 'topaz' (premium quality)
prompt	string	no	Optional prompt to guide the upscaling (only for clarity model)
creativity	number	no	Creativity level 0-1 for clarity model (higher = more creative enhancement)

Parameter	Type	Required	Notes
image_url	string	yes	URL of the image with subject to keep
prompt	string	no	Text description of the new background
ref_image_url	string	no	URL of a reference image for the new background style
model		yes	Background replacement model. The default, ideogram-keep, keeps the product's own pixels; the others can redraw the product

Parameter	Type	Required	Notes
image_url	string	yes	URL of the image to edit
mask_url	string	yes	URL of a mask image where white areas indicate objects to remove
prompt	string	no	Optional description of what fills the masked area. Without it, the masked objects are removed.

Parameter	Type	Required	Notes
prompt	string	yes	Text description of the image to generate
model	string	no	Generation model. Nano Banana 2 Lite is the default
aspect_ratio	string	no	Output aspect ratio
num_images	integer	no	Number of images to generate (1-4)

Parameter	Type	Required	Notes
image_url	string	yes	URL of the source image to transform
prompt	string	yes	Description of the desired transformation
strength	number	no	Transformation strength 0-1 (higher = more change from original)
model	string	no	Transformation model. Nano Banana 2 Lite is the default

Parameter	Type	Required	Notes
image_url	string	yes	URL of the source image whose visible text should be translated
target_language	string	yes	Target language code or name for the translated image text
source_language	string	no	Optional source language code or name
model	string	no	Image translation model. Nano Banana 2 Lite is the default

Parameter	Type	Required	Notes
prompt	string	yes	Description of the video to generate
image_url	string	no	Starting image. Required for veo31, ltx and kling. Optional for h3-max. Not used by veo31ref
image_urls	array	no	Reference images (1-5) that keep the subject the same. Required for veo31ref, not used by other models
model	string	no	Video generation model
duration	string	no	Video duration in seconds (e.g., '4', '6', '8')
resolution	string	no	Video resolution
aspect_ratio	string	no	Video aspect ratio
generate_audio	boolean	no	Whether to generate audio for the video

Parameter	Type	Required	Notes
video_url	string	yes	URL of the source video to upscale
model	string	no	Video upscale model
scale		no	Upscale factor
target_resolution	string	no	Target output resolution
source_width	integer	yes	Source video width in pixels
source_height	integer	yes	Source video height in pixels
source_duration_seconds	number	yes	Source video duration in seconds
source_frame_rate	number	no	Source video frame rate

Parameter	Type	Required	Notes
image_url	string	yes	URL of the image to describe
detail_level	string	no	Level of detail: 'caption' (brief), 'detailed-caption' (standard), 'more-detailed-caption' (comprehensive)

Parameter	Type	Required	Notes
image_url	string	yes	URL of the image to analyze

Parameter	Type	Required	Notes
image_url	string	yes	URL of the product image (ideally with transparent background)
scene_description	string	no	Description of the lifestyle scene to place the product in
ref_image_url	string	no	URL of a reference image for the scene style
model	string	no	Lifestyle image model. The default, ideogram-keep, keeps the product's own pixels; the others can redraw the product
placement_type	string	no	How to place the product in the scene
position	string	no	Position of the product in the scene (for manual_placement)
num_results	integer	no	Number of variations to generate

Parameter	Type	Required	Notes
person_image_url	string	yes	URL of the person/model image
garment_image_url	string	yes	URL of the clothing item image

Parameter	Type	Required	Notes
image_url	string	yes	URL of the image to convert to 3D
model	string	no	3D generation model
topology	string	no	Mesh topology type
target_polycount	integer	no	Target polygon count for the mesh
enable_pbr	boolean	no	Enable PBR (physically-based rendering) textures

Parameter	Type	Required	Notes
glb_url	string	yes	HTTPS URL of a binary .glb file up to 100 MB
simplify_ratio	number	no	Target mesh simplification ratio; 1 keeps full geometry
simplify_error	number	no	Maximum simplification error tolerance
dedup	boolean	no	Remove duplicate vertices, meshes, textures, and materials
prune	boolean	no	Remove unused data from the model
draco	boolean	no	Apply Draco mesh compression
instance	boolean	no	Instance repeated meshes
flatten	boolean	no	Flatten the scene hierarchy
join	boolean	no	Join separate meshes where possible
weld	boolean	no	Weld nearby duplicate vertices
quantize	boolean	no	Quantize vertex attributes before compression
texture_format	string	no	Optional texture output format
texture_size		no	Optional maximum texture dimension

Remove backgrounds from multiple images in a batch operation. Processes multiple images at once for efficiency. Returns results for each image including success/failure status. Args: - images (array, required): Array of {id, image_url} objects - model (string): Model to use for all images…

Parameter	Type	Required	Notes
images	array	yes	Array of images to process
model	string	no	Model to use for all images

Upscale multiple images in a batch operation. Processes multiple images at once for efficiency. Returns results for each image. Args: - images (array, required): Array of {id, image_url} objects - scale (number): Upscale factor 2-4x (default: 2) - model (string): Upscaler model (default:…

Parameter	Type	Required	Notes
images	array	yes	Array of images to upscale
scale	integer	no	Upscale factor for all images
model	string	no	Upscaler model

No parameter schema on file.

Free

$0 forever

Every tool, no call limit
No card, and mcprush charges nothing for it
Source published under MIT
Runs on your machine — nothing of it reaches our gateway
Nothing to cap, because mcprush bills nothing

Nothing here is billable: mcprush charges nothing to install Sirv AI Studio or to call it, at any volume.

ATrust score 93 / 100

Release 1.1.1 · scanned 5 Nov 2026

Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.

Threat score100

Client score93

Findings1 threat · 0 capability

Capabilityhigh

Coveragesource

Verificationrepo

Surface digestc73b69f99295ea777ea2023c37abb1a781c5c523e99ffa78e6307a39d4cafde6

Score vectorthreat 100 → client 93

Term	Level	What it prices	Points
capability-exposure	high	capability blast radius (high) — client exposure if the model is manipulated	−6
verification-discount	repo	publisher verification (public source) — no provenance, but the source is public and inspectable	−1
coverage-honesty	source	inspection depth (source) — how much of the target the scan could see	−0

score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.

What the scan could actually read

A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.

Tool surfaceread

Implementation sourceread

Package metadataread

Live transportnot available

Tools were statically extracted from the published source (53 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".

Capability — what it could do if the model were manipulated

Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.

Tool	Capability tags	Why the tag was assigned
sirv_get_usage	no tags
sirv_list_organizations	no tags
sirv_alt_text	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_depth_map	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_list_assets	no tags
sirv_get_asset	no tags
sirv_search_assets	no tags
sirv_find_similar_assets	no tags
sirv_update_asset	state-write	state-write: declares readOnlyHint: false, destructiveHint: true
sirv_delete_asset	state-write	state-write: declares readOnlyHint: false, destructiveHint: true
sirv_get_asset_metadata	no tags
sirv_update_asset_metadata	state-write	state-write: declares readOnlyHint: false, destructiveHint: true
sirv_generate_asset_alt_text	state-write	state-write: declares readOnlyHint: false
sirv_bulk_asset_operation	state-write	state-write: declares readOnlyHint: false, destructiveHint: true
sirv_import_asset_url	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_batch_remove_background	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_batch_upscale	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_generate	state-write	state-write: declares readOnlyHint: false
sirv_image_to_image	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_image_translation	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_video_generation	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_video_upscale	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_remove_background	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_upscale	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_background_replace	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_object_removal	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_list_products	no tags
sirv_get_product	no tags
sirv_create_product	state-write	state-write: declares readOnlyHint: false
sirv_update_product	state-write	state-write: declares readOnlyHint: false, destructiveHint: true
sirv_delete_product	state-write	state-write: declares readOnlyHint: false, destructiveHint: true
sirv_search_products	no tags
sirv_list_products_with_assets	no tags
sirv_link_assets_to_product	state-write	state-write: declares readOnlyHint: false
sirv_bulk_product_action	state-write	state-write: declares readOnlyHint: false, destructiveHint: true
sirv_product_lifestyle	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_virtual_try_on	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_image_to_3d	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_optimize_glb	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_shopify_status	no tags
sirv_push_image_to_shopify	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_sync_shopify_products	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_cdn_search_files	no tags
sirv_cdn_get_file_metadata	no tags
sirv_cdn_manage_folder	no tags
sirv_cdn_upload_from_url	no tags
sirv_cdn_update_file_metadata	no tags
sirv_cdn_manage_file	no tags
sirv_get_supplier_sftp_status	no tags
sirv_get_supplier_review_assist_requirements	no tags
sirv_run_supplier_autofix	external-sink	external-sink: declares readOnlyHint: false, openWorldHint: true
sirv_list_workflows	no tags
sirv_get_workflow	no tags

Toxic-flow graph

The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.

The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.

Supply chain and provenance

Integrityfirst-seen

Surface digestc73b69f99295ea77

Scanned2026-11-05

This is the first scan of this surface here, so there is nothing yet to compare it against.

Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.