SurgePix

Home>Skills Library>Generate Brand Placement

Brand Placement Lock

Generate images while locking a brand logo and/or fixed text to a stable position and scale, so every output under the same lock profile keeps those elements in the same place. SurgePix Brand Placement locks your logo and tagline to a fixed 9-grid anchor, so every AI-generated poster, Amazon lifestyle image, ecommerce banner, and brand marketing campaign stays perfectly on-brand. Upload your transparent PNG logo and fixed text once, then generate on-brand variations in bulk with identical logo size, margin, and position. Ideal for Amazon Storefront branding, Shopify series posters, Instagram branding, and franchise brand marketing. No drifting, no rewriting, no covering — just scalable branding consistency for ecommerce growth.

npx skills add https://github.com/surgepix/agent-skills --skill surgepix-generate-brand-placement

Download ZIP GitHub

SurgePix Fixed Text Position Across Generations

This skill is about positional consistency of locked brand elements, not about inventing a new logo. To create a logo first, use surgepix-generate-logo.

What the Skill Does

Action	Description
Lock logo position	Place a provided logo at `--logo-anchor`; size and margin follow platform defaults
Lock fixed text	Place brand text / slogan at `--text-anchor` — never rewrite the locked string
Choose output format	Logo and fixed text each have their own nine-grid anchor; they may share a zone or sit in different corners
Keep layout stable	With the same lock flags, batch outputs and re-runs keep locked elements at the same relative position and scale
Free the rest	Only the unlocked area (scene / subject / background) may change per `--prompt`
Set output count	Output piece count is driven by `--count` (default `1`, range `1-9`) and the series intent in `--prompt` — not by how many `--reference` files are uploaded
Reference images	Optional style / mood / composition references only — not one-to-one edit bases; reference count does not change output count
Set output size	Set a uniform output size via `--size`; when omitted, use the API default (references do not set per-piece sizes)
Check task status	When run with `--nowait true`, look up via `taskId` using surgepix-query-task
Download result	Returns a single image URL for one piece; a ZIP containing all pieces for multiple
By default (`--nowait false`) the API waits for completion and returns the final `download` link directly. Pass `--nowait true` to get a `taskId` immediately; then check progress via the surgepix-query-task skill.

Workflow

Step 0: Check environment (required)
Before running, verify config:
```
node "<skills-dir>/surgepix-setup/scripts/check_env.mjs"
```
- Exit 0 → proceed to Step 1
- Exit 1 → configure .env following the surgepix-setup skill, then retry

Step 1: Gather inputs

Input	Required?	Description
`--prompt`	Required	Unlocked scene guidance; do not describe placement
`--logo`	Conditionally required	At least one of `--logo` / `--fixed-text`; local path or URL; prefer transparent PNG
`--fixed-text`	Conditionally required	At least one of `--logo` / `--fixed-text`; exact fixed text string to lock
`--reference`	Optional	Style / mood / composition reference(s); repeatable; does not set output count; `JPEG` / `JPG` / `PNG` / `WEBP`, max 20 MB each
`--logo-anchor`	Optional	Logo nine-grid anchor when `--logo` is set; default `top-right`
`--text-anchor`	Optional	Fixed-text nine-grid anchor when `--fixed-text` is set; default `top-right`
`--count`	Optional	Output piece count, `1-9`, default `1`; align with series intent in `--prompt`
`--size`	Optional	Uniform `WxH`; omitted -> API default
`--session-id`	Optional	Frontend session grouping only; not a regenerate switch
`--nowait`	Optional	Default `false`

Step 2: Run generate-brand-placement

node "<skills-dir>/surgepix-lock-brand-mark-position/scripts/lock_brand_mark_position.mjs" \
  --prompt "<scene guidance>" \
  [--logo "<path-or-url>"] [--logo-anchor <anchor>] \
  [--fixed-text "<text>"] [--text-anchor <anchor>] \
  [--reference "<path-or-url>" ...] [--count <1-9>] \
  [--size <WxH>] [--session-id <id>] [--nowait <true|false>]

Flag	Required?	Description
`--prompt <text>`	Required	Scene guidance
`--logo <path-or-url>`	Conditionally required	Logo to lock; at least one of `--logo` / `--fixed-text`
`--logo-anchor <value>`	Optional	Logo nine-grid; default `--top-right` (ignored if no `--logo`)
`--fixed-text <text>`	Conditionally required	Exact fixed text; at least one of `--logo` / `--fixed-text`
`--text-anchor <value>`	Optional	Fixed-text nine-grid; default `top-right` (ignored if no `--fixed-text`)
`--reference <path-or-url>`	Optional	Style reference(s); repeatable; does not change output count
`--count <1-9>`	Optional	Output piece count; default `1` driven by count + prompt intent, not reference count
`--size <WxH>`	Optional	Uniform output size; omitted -> API default
`--session-id <id>`	Optional	Numeric `sessionId`; frontend session grouping only - not for regeneration
`--nowait <true\|false>`	Optional	Default `false` ; `true` returns `taskId`

Step 3: Parse output

Sync success (single):

{"ok":true,"taskId":"task_abc123","sessionId":123,"progress":"succeeded","download":"https://...","imageCount":1,"resultType":"zip","note":"API 仅返回 ZIP 下载地址，不含单张图片 URL；禁止编造单张链接"}

Async submitted (ZIP):

{"ok":true,"taskId":"task_abc123","sessionId":123,"progress":"succeeded","download":"https://.../xxx.zip","imageCount":4,"resultType":"zip"}

Async submitted:

{"ok":true,"async":true,"taskId":"task_abc123","sessionId":123,"progress":"processing","download":null,"hint":"..."}

Failure:

{"ok":false,"error":"..."}

Step 4: Present result
- On success: Show download; for multiple pieces show only the ZIP; sessionId is optional to show (frontend grouping)
- On failure: Report the error field. Common causes: missing_required_param / unsupported_image_format / image_too_large / invalid_param_value / generation_failed
- When the user is not satisfied:
  - Local changes -> switch to image-edit
  - Full redo -> update --prompt and lock-related flags / --count and re-run this skill
  - Do not treat --session-id as a regenerate mechanism