Omniscient Master Suite OMNISCIENT · 3.1
Sign in with OmniumPortalSign in

Image Generator v1.0.0

AI & ML image-generator 3 actions MCP Custom UI

Full dual-provider batch image generator (OpenAI gpt-image-1 and Google Gemini 'Nano Banana'): provider switch, single-prompt and JSON-driven batch modes (load a file or paste JSON, extract by explicit keys or grab every string, three ways to combine with a base prompt), continue-on-error, live progress + log, preview and save, a Help dialog, and settings that autosave every second. API keys are read from the secure vault, never stored in plaintext.

Open in Omniscient Back to apps

The omniscient:// link needs a protocol handler on your machine, which is not built yet. Until then it does nothing — use the commands below.

Actions

Jump to: generategenerate_batchcheck_key

Generate one image

generate

Generate a single image from one prompt with the chosen provider and save it to a folder - the same single-prompt path the GUI's Generate button runs.

Inputs

NameTypeRequired / defaultValues and notes
prompt text required
provider enum optional, default openai openaigemini
size enum optional, default 1024x1024 1024x10241024x15361536x1024
quality enum optional, default high highmediumlow
background enum optional, default auto autotransparentopaque
format enum optional, default png pngjpegwebp
n int optional, default 1
model enum optional, default gemini-2.5-flash-image gemini-3.1-flash-image-previewgemini-3-pro-image-previewgemini-2.5-flash-image
out_dir path optional, default ""

Outputs

NameTypeNotes
file path
provider string

Generate a JSON batch

generate_batch

Generate one image per prompt from a JSON file and/or inline JSON - the same JSON-keys/ignore-keys extraction and concatenation-mode logic the GUI's batch mode uses - saving every image to an output folder.

Inputs

NameTypeRequired / defaultValues and notes
prompts_file path optional, default ""
prompts json optional, default []
keys string optional, default ""
ignore_keys bool optional, default false
base_prompt text optional, default ""
concat_mode enum optional, default No Concatenate No ConcatenatePrepend JSON to Base PromptAppend JSON to Base Prompt
provider enum optional, default openai openaigemini
size enum optional, default 1024x1024 1024x10241024x15361536x1024
quality enum optional, default high highmediumlow
background enum optional, default auto autotransparentopaque
format enum optional, default png pngjpegwebp
n int optional, default 1
model enum optional, default gemini-2.5-flash-image gemini-3.1-flash-image-previewgemini-3-pro-image-previewgemini-2.5-flash-image
out_dir path optional, default ""
continue_on_error bool optional, default true

Outputs

NameTypeNotes
count int
failed int
out_dir path

Check API key

check_key

Verify that the stored API key for a provider is present and accepted - the same check OpenAI's 'Verify Key / Model' button runs; adds an equivalent lightweight check for Gemini, which the original app never had.

Inputs

NameTypeRequired / defaultValues and notes
provider enum optional, default openai openaigemini

Outputs

NameTypeNotes
ok bool
detail string

Use it

Generate one image

Command line

shell
python app.py --action generate --params '{"prompt": "<prompt>"}'

Run from the app’s own folder. Single quotes work in bash and PowerShell; in cmd.exe use double quotes and escape the inner ones.

MCP tool

mcp
run_action("image-generator", "generate", {"prompt": "<prompt>"})

Workflow node

json
{
  "id": "n0",
  "app": "image-generator",
  "action": "generate",
  "params": {
    "prompt": "<prompt>"
  },
  "inputs": {}
}

Generate a JSON batch

Command line

shell
python app.py --action generate_batch --params '{}'

Run from the app’s own folder. Single quotes work in bash and PowerShell; in cmd.exe use double quotes and escape the inner ones.

MCP tool

mcp
run_action("image-generator", "generate_batch", {})

Workflow node

json
{
  "id": "n0",
  "app": "image-generator",
  "action": "generate_batch",
  "params": {},
  "inputs": {}
}

Check API key

Command line

shell
python app.py --action check_key --params '{}'

Run from the app’s own folder. Single quotes work in bash and PowerShell; in cmd.exe use double quotes and escape the inner ones.

MCP tool

mcp
run_action("image-generator", "check_key", {})

Workflow node

json
{
  "id": "n0",
  "app": "image-generator",
  "action": "check_key",
  "params": {},
  "inputs": {}
}

Example prompts

Things you can say to Claude, Cursor or Codex once the MCP server is registered.