seeany-sun-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@seeany-sun-mcp Generate a product image of a white sneaker on a marble surface"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
SeeAny Sun MCP
This directory provides a local stdio MCP Server that can be called by MCP clients such as Codex, Claude Desktop, and Cursor.
It supports two modes:
demo: no API key required; uses product images in the repository for offline simulation and does not call paid APIs.live: after settingSEEANY_API_KEY, calls the main sitehttps://api.seeany.com; supports image upload, AI creation, task polling, download, and AI refinement.
Authentication environment variables
Use uniformly:
SEEANY_API_KEY=sk-sa-xxxxxxxxDo not write the real Key into Git, browser code, or chat messages. Copy .env.example to your local environment and fill it in. If there is no Key, the service automatically enters Demo mode; you can also use SEEANY_MCP_MODE=demo to force offline mode.
Related MCP server: metalevel/snapix-mcp-server
Local installation and testing
cd mcp-server
npm install
npm run build
npm run smokeThe Smoke test forces Demo mode and will not incur real charges.
If you need to perform a real end-to-end acceptance test (which will incur charges on the main account), after confirming that the new Key has been set, run:
npm run live-smokeConnecting Codex
codex mcp add seeany-sun-mcp -- npx --yes seeany-sun-mcpInstall as a Codex plugin (MCP + Skill)
The repository also provides a Codex plugin manifest: after installation, you will get both the SeeAny MCP tools and the seeany-product-visuals product visual workflow Skill.
codex plugin marketplace add juin20260102-oss/seeany-sun-mcp --sparse .agents/plugins
codex plugin add seeany-sun-mcp@seeanyAfter installation, start a new Codex conversation; for real image generation, you still need to set SEEANY_API_KEY in your local environment. If you only need the MCP tools, use the codex mcp add command above.
If you use the real API, set SEEANY_API_KEY in the same environment where you start the MCP. After connecting, it is recommended to call seeany_get_capabilities first, then call in the following order:
seeany_upload_asset(when there is a reference image)seeany_refine_prompt(optional)seeany_generate_product_imageseeany_wait_generationseeany_download_assets
MCP Inspector
npm run inspectCurrent real API mapping
MCP tool | SeeAny API |
|
|
|
|
|
|
|
|
Tasks default to asynchronous mode. The main site API's image/video tasks are billed according to account configuration; seeany_quote_generation only returns an estimate and does not pre-deduct charges.
SeeAny Skill
The repository includes skills/seeany-product-visuals, used to guide the Agent through the complete flow of “upload reference image → prompt refinement → quote → generate → poll → download” when calling the MCP. The npm package also includes this Skill directory.
npm package name
The current package name is set to seeany-sun-mcp, suitable for first publishing a public unscoped package. Before the official release, you need to confirm the name is available on npm; if SeeAny has established an npm scope, you can also migrate to @seeany/sun-mcp.
The simplest invocation after publishing:
npx --yes seeany-sun-mcpCheck the package contents before publishing:
npm run pack:checkAutomated testing and publishing
GitHub Actions will run on Push and Pull Request to the main branch using Node.js 20, 22, and 24 to perform plugin validation, MCP Smoke Test, and npm packaging checks.
When publishing a new version, first update the versions in package.json, package-lock.json, and .codex-plugin/plugin.json in sync, then create a Git Tag with the same version:
git tag v0.5.0
git push origin v0.5.0release.yml verifies that the Tag matches the package version, then publishes via npm Trusted Publishing and creates a GitHub Release. Before first use, you need to add the GitHub Actions Trusted Publisher in the npm package settings:
GitHub user:
juin20260102-ossRepository:
seeany-sun-mcpWorkflow:
release.ymlAllowed action:
npm publish
Available Tools
11 toolsseeany_create_product_visualA
Execute a previously quoted product-visual plan, wait for its stored job, and download outputs. In live mode confirm_cost must be true. Reusing a plan_id resumes its recorded job instead of submitting another one.
Inputs: plan_id (required), confirm_cost (required), output_dir, timeout_seconds.
seeany_download_assetsA
Download completed SeeAny output images into a local directory and return their file paths.
Inputs: job_id (required), output_dir (required).
seeany_generate_product_imageB
Create an asynchronous SeeAny product-image task, then use seeany_wait_generation to receive results.
Inputs: prompt (required), count, model, use_case, +3 more.
seeany_get_accountARead-only
Show whether this MCP is using the local demo or the authenticated SeeAny API.
Inputs: no named parameters.
seeany_get_capabilitiesARead-only
List SeeAny image generation models, ratios, resolutions, and current execution mode.
Inputs: no named parameters.
seeany_get_generationARead-only
Get the latest status, progress, and output URLs for a SeeAny generation task.
Inputs: job_id (required).
seeany_plan_product_visualARead-only
Plan a complete product-visual job and estimate its cost without creating a paid generation task. Returns a plan_id for explicit execution.
Inputs: request (required), count, model, use_case, +4 more.
seeany_quote_generationARead-only
Estimate generation cost before creating a task. This is an estimate and does not reserve credits.
Inputs: count (required), model (required), use_case (required), resolution.
seeany_refine_promptB
Use SeeAny AI prompt refinement to turn a rough product-image request into a production-ready prompt.
Inputs: prompt (required), action, extra_context, reference_asset_ids.
seeany_upload_assetA
Upload a local product image to SeeAny and return an asset id and URL for later generation.
Inputs: file_path (required).
seeany_wait_generationARead-only
Poll a SeeAny generation task until it reaches a terminal state or the timeout expires.
Inputs: job_id (required), timeout_seconds.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 11 tools
Several tools overlap: seeany_plan_product_visual and seeany_quote_generation both estimate costs, while seeany_create_product_visual and seeany_generate_product_image both create generation jobs and handle waiting/downloading. The separate 'visual' and 'image' workflows are not clearly distinguished by descriptions, so an agent could easily pick the wrong one.
All tools share the seeany_ prefix and follow a consistent verb_noun snake_case pattern (get_account, upload_asset, generate_product_image, wait_generation). There are no mixed conventions or unpredictable names.
At 11 tools, the server is within a reasonable scope for an image-generation API client. It could be trimmed by consolidating overlapping quote/plan and create/generate tools, but the count is not excessive.
The core lifecycle is covered: asset upload, prompt refinement, cost estimation, job creation, status polling, waiting, and downloading outputs. Missing cancellation or asset listing are minor gaps, not dead ends.
Maintenance
Related MCP Connectors
MCP server for ByteDance Seedream AI image generation
MCP server for Wan AI video generation
MCP server for Qwen Image 3 AI image generation
MCP server for Product Management
Related MCP Servers
- AlicenseAqualityDmaintenanceMCP server for multi-provider AI image generation (AWS Bedrock, OpenAI, Google Gemini) enabling image generation, transformation, and editing through a unified interface.41MIT
- AlicenseNot gradedqualityDmaintenanceMCP server for image optimization, conversion, AI generation, and gallery management via SnapiX API.19 npmMIT
- FlicenseAqualityBmaintenanceMCP server for the company image2 text-to-image API, enabling AI tools to generate images directly from text prompts.2-
- AlicenseAqualityCmaintenanceMCP server for OfflineCreator Studio that enables image and video generation, including model listing, credit management, generation tasks, and output downloads.999 npmMIT