Publish Sign in

Ecommerce Image Workflow

Reference-product ecommerce image workflow for generating a compact set of product-faithful main, feature, and lifestyle images from real product reference photos. V1 requires uploaded product…

You say
Install this skill Read the source first Free Written by nexu-io · unverified publisher
Context cost
6.2k tokensestimated from the bundle, loaded when it triggers
Bundle
3 files · 24.9 kBtext throughout, nothing executable
Licence
Apache-2.0free to use
Last change
—no release on file
Servers it uses
Noneruns standalone

What it does

Reference-product ecommerce image workflow for generating a compact set of product-faithful main, feature, and lifestyle images from real product reference photos. V1 requires uploaded product imagery and intentionally defers brief-only concept generation and platform-specific batch exports.

Installed, it changes the agent in these ways.

What this skill changes about the agent is not written down here yet. The listing was collected from its source, and the description is in its own SKILL.md.

Workflow

Runs a procedure end to end.

e-commerce
Filed under

Finance & Commerce

The skill itself

This is the whole product. A skill is instructions the model reads, so there is nothing behind the listing you cannot see first — the front matter loads with every session, and the body below it loads when the skill triggers.

SKILL.md9.7 kB · 298 lines
--- name: ecommerce-image-workflow en_name: "Ecommerce Image Workflow" description: | Reference-product ecommerce image workflow for generating a compact set of product-faithful main, feature, and lifestyle images from real product reference photos. V1 requires uploaded product imagery and intentionally defers brief-only concept generation and platform-specific batch exports. triggers: - "ecommerce product images" - "product image set" - "product photography workflow" - "product main image" - "product feature shot" - "reference product commerce images" - "lifestyle product image" - "amazon product images" - "shopify product images" - "taobao product images" od: mode: image surface: image category: image-generation scenario: marketing preview: type: html entry: example.html design_system: requires: false example_prompt: | Use the Ecommerce Image Workflow to turn my uploaded product reference photo into a compact ecommerce image set: one main packshot, one feature highlight image, and one lifestyle scene. Preserve the exact product identity, color, material, logo placement, structure, and proportions. ---
37# Ecommerce Image Workflow
38
39Create a compact ecommerce image set from real product reference imagery.
40This V1 skill is intentionally narrow: it supports **reference-product mode
41only**. If the user only describes a product and does not provide a product
42photo, ask for one and stop. Do not create a brief-only concept product in
43this version.
44
45## Resource map
46
47```text
48ecommerce-image-workflow/
49|-- SKILL.md
50|-- example.html
51`-- references/
52 `-- checklist.md
53```
54
55## What this skill produces
56
57By default, generate three ecommerce-ready image assets for one product:
58
591. **Main image** - clean product-first packshot on white or soft neutral
60 background.
612. **Feature image** - one selling point shown clearly with controlled callout
62 space, without relying on tiny unreadable in-image text.
633. **Lifestyle image** - product shown in a plausible use context while keeping
64 the product faithful to the reference.
65
66Also create:
67
68- image-manifest.json describing reference inputs, slots, prompts, outputs,
69 aspect ratios, and fidelity notes.
70- ecommerce-gallery.html as a small preview gallery linking the generated
71 files and summarizing the image roles.
72
73## Input contract
74
75Required:
76
77- At least one uploaded product reference image in the active project.
78
79Ask only for missing essentials:
80
81- Product name or short label if it is not obvious.
82- Main selling point if the feature image cannot be inferred safely.
83- Target marketplace or aspect only if the user asks for platform-specific
84 framing.
85
86Do not ask broad discovery questions. Keep the workflow moving.
87
88## Workflow
89
90### Step 0 - Confirm reference-product mode
91
92Before planning, verify that the current project includes a real product
93reference image.
94
95If no product image is available, reply:
96
97> Please upload at least one product reference image first. This V1 workflow
98> preserves a real product from reference photos; brief-only concept generation
99> is deferred to a later version.
100
101Then stop.
102
103### Step 1 - Extract product identity anchors
104
105Inspect the reference image and write a short internal identity lock:
106
107- Product category and form factor.
108- Shape and silhouette.
109- Primary colors and materials.
110- Logo, label, pattern, fasteners, ports, straps, handles, or other fixed
111 details.
112- Scale cues and proportions.
113- What must not change.
114
115Use these anchors in every generation prompt.
116
117### Step 2 - Build a three-slot shot plan
118
119Create a compact shot plan before dispatch:
120
121| Slot | Default aspect | Goal |
122|---|---:|---|
123| main | 1:1 | Product-first marketplace image on white or soft neutral background |
124| feature | 4:5 | One clear selling point with close-up detail or simple callout space |
125| lifestyle | 4:5 | Realistic use context with the product still visually faithful |
126
127If the project metadata provides imageAspect, use it when the user expects a
128single aspect across the set. Otherwise use the slot defaults above.
129
130### Step 3 - Compose prompts with a fidelity lock
131
132Every prompt must include this product fidelity instruction near the top:
133
134```text
135Preserve the exact product identity from the reference image: shape,
136silhouette, color, material, logo/label placement, visible construction
137details, and proportions. Do not redesign the product. Do not add, remove,
138or relocate product features.
139```
140
141Then add slot-specific instructions:
142
143#### Main image prompt
144
145- Product centered and fully visible.
146- White, off-white, or very light grey background.
147- Soft studio lighting with clean shadow.
148- No props unless the user asked for them.
149- No in-frame marketing text.
150
151#### Feature image prompt
152
153- Focus on one user-provided or safely inferred feature.
154- Use close-up composition, cutaway-style crop, or clean negative space for
155 later designer-added labels.
156- Keep the product visually balanced in the frame. If no explicit callout
157 structure is being generated, center the product. If label space is needed,
158 offset the product only slightly and make the empty space feel intentional.
159- Do not invent certifications, performance numbers, materials, or claims.
160- Avoid tiny rendered text; leave label space instead.
161
162#### Lifestyle image prompt
163
164- Use a realistic environment matched to the product category.
165- Keep the product the focal point.
166- Show human interaction only if it helps explain use and does not obscure the
167 product.
168- Preserve product scale and structure.
169
170### Step 4 - Dispatch through the media contract
171
172Use the unified OpenDesign media dispatcher. Do not call provider APIs or
173custom model commands directly.
174
175For each slot, run the standard generate/wait loop:
176
177```bash
178# POSIX bash. Do not call provider APIs directly.
179out=$("$OD_NODE_BIN" "$OD_BIN" media generate \
180 --project "$OD_PROJECT_ID" \
181 --surface image \
182 --model "<imageModel from metadata>" \
183 --aspect "<slot aspect or imageAspect from metadata>" \
184 --image "<project-relative product reference image>" \
185 --output "<product-slug>-<slot>.png" \
186 --prompt "<full slot prompt>")
187ec=$?
188if [ "$ec" -ne 0 ]; then echo "$out" >&2; exit "$ec"; fi
189
190last=$(printf '%s\n' "$out" | tail -1)
191task_id=$(printf '%s\n' "$last" |
192 python3 -c "import sys,json; d=json.load(sys.stdin); print(d.get('taskId',''))" 2>/dev/null)
193since=$(printf '%s\n' "$last" |
194 python3 -c "import sys,json; d=json.load(sys.stdin); print(d.get('nextSince',0))" 2>/dev/null)
195since="${since:-0}"
196
197while [ -n "$task_id" ]; do
198 out=$("$OD_NODE_BIN" "$OD_BIN" media wait "$task_id" --since "$since")
199 ec=$?
200 last=$(printf '%s\n' "$out" | tail -1)
201 since=$(printf '%s\n' "$last" |
202 python3 -c "import sys,json; d=json.load(sys.stdin); print(d.get('nextSince',0))" 2>/dev/null)
203 since="${since:-0}"
204 if [ "$ec" -eq 0 ]; then
205 task_id=""
206 elif [ "$ec" -ne 2 ]; then
207 echo "$out" >&2
208 exit "$ec"
209 fi
210done
211
212printf '%s\n' "$last"
213```
214
215The final line must be JSON with {"file": {"name": "...", ...}}.
216Record each final returned filename in image-manifest.json.
217
218If the active image model or provider cannot use --image, stop and tell the
219user that this workflow needs a reference-capable image generation path for
220product fidelity.
221
222### Step 5 - Write image-manifest.json
223
224After generation, create a project file named image-manifest.json:
225
226```json
227{
228 "workflow": "ecommerce-image-workflow",
229 "mode": "reference-product",
230 "productName": "Example product",
231 "referenceImages": ["reference-product.png"],
232 "fidelityNotes": [
233 "Preserve product identity, color, material, construction, and proportions.",
234 "Do not treat these outputs as platform-compliance proof without human review."
235 ],
236 "slots": [
237 {
238 "id": "main",
239 "role": "marketplace packshot",
240 "aspect": "1:1",
241 "output": "example-product-main.png",
242 "promptSummary": "Centered product-first packshot on a clean neutral background."
243 },
244 {
245 "id": "feature",
246 "role": "single feature highlight",
247 "aspect": "4:5",
248 "output": "example-product-feature.png",
249 "promptSummary": "Close-up or negative-space composition for one verified selling point."
250 },
251 {
252 "id": "lifestyle",
253 "role": "usage context",
254 "aspect": "4:5",
255 "output": "example-product-lifestyle.png",
256 "promptSummary": "Realistic scene with the product as the focal point."
257 }
258 ]
259}
260```
261
262Keep the manifest honest. If a detail is unknown, write null or a short note
263instead of inventing claims.
264
265### Step 6 - Write ecommerce-gallery.html
266
267Create a simple single-file HTML gallery that:
268
269- Shows the reference image first.
270- Shows the three generated slots with their role names.
271- Lists product-fidelity notes.
272- Links to image-manifest.json.
273- Uses system fonts and local project files only; no CDN imports.
274
275### Step 7 - Hand off
276
277Reply with:
278
279- The generated filenames.
280- A one-sentence summary of the fidelity lock used.
281- A reminder that marketplace-specific compliance, final text overlays, and
282 claim/legal review remain human review steps.
283
284Do not emit an <artifact> tag.
285
286## Hard rules
287
288- V1 requires real product reference imagery. No brief-only concept products.
289- One product per run.
290- Default to exactly three slots: main, feature, lifestyle.
291- Preserve the product; do not redesign it.
292- Do not invent claims, certifications, measurements, ingredients, or
293 performance data.
294- Use "$OD_NODE_BIN" "$OD_BIN" media generate; do not call provider APIs
295 directly.
296- Always create image-manifest.json after generation.
297- Run references/checklist.md before handoff.
298
In the file
SKILL.md1,350 words
Files3
LicenceApache-2.0
Why you can read it

Nothing in a skill executes. The client loads the text and the model follows it, so a skill can be audited the way a runbook is — by reading it.

What it costs in context

Skills are not billed by the call. They are paid for in context: every token the instructions occupy is a token your code, your diff and your conversation cannot use. Here is what this one takes and when it takes it.

≈300
always loaded
The name and description, so the model knows the skill exists and when to reach for it.
5,925
on trigger
The instruction body and 2 supporting files, read only when the skill fires.
3.1%
of a 200k window
Ten skills this size would take about 31% of the window before you open a file.
050k100k150k200k context window

6.2k tokens, estimated from the bundle at four bytes to the token, held for the rest of the session once it triggers. Heavy. Teams tend to install this one per project rather than globally, and load it only when the job comes up.

Servers bill, skills cost

A server charges by the month. A skill charges once per session, in context, and then keeps charging it for as long as the session lives.

Before and after

The same question, put to the same model twice: once as it comes, and once with these instructions loaded.

No worked example has been published for this skill yet.

Adoption
Installsnone yet
Ratingno reviews yet

The procedure it runs

The procedure has not been published here. It is in the skill’s own SKILL.md, which its author has not sent to the marketplace yet.

Prose, not code

These steps are written for a model to follow, not executed by a runtime. It can still be told to skip one, and it will say so when it does.

Servers it uses

None. This skill calls no MCP servers at all.

Everything it needs is in the instructions, so it works in a project with nothing connected — the model reads the file and changes how it works with what it can already reach.

It writes no files and reaches no network. All it changes is how the model reasons and writes.

What it asks for
Writes filesno
Network accessno

Read from the allowed-tools line of this skill’s own SKILL.md. A skill grants no permissions of its own — it can only ask for tools your client already has.

What it will not do

Every skill is narrow, and the useful ones say where they stop. These are the jobs this one is the wrong tool for.

What this skill is not for has not been published here. Nothing is implied by that: it is a section the author has not filled in.

What is in the bundle

3 files, 24.9 kB on disk. A bundle is text throughout: the instructions the model reads, plus the templates it fills in.

  • SKILL.md9.7 kB
  • example.html12.0 kB
  • references/checklist.md3.2 kB
What is not in it

No dependencies and nothing executable: a skill is text the agent reads, so the bundle is 3 files you can review in full before installing. The Apache-2.0 licence covers the templates and examples as well as the instructions.

Install

Installing copies the bundle into your project. Nothing runs at install time — the files sit on disk until the model reads them.

# Ecommerce Image Workflow · 6.2k tokens when loaded npx mcprush@latest skill add nexu-io/ecommerce-image-workflow

Writes to .claude/skills/ecommerce-image-workflow/ in the current project. Add --global to put it in your home directory instead, for every project.

Which clients pick it up on their own

A skill is a folder of text. A client with a skills folder reads it without being told; everywhere else the same text works, it is just handed to the model rather than found.

Claude Code.claude/skills/
Claude Desktop
ChatGPT
Cursor.cursor/skills/
VS Code.github/skills/
Codex CLI.agents/skills/
Gemini CLI.gemini/skills/
Grok.grok/skills/
Zed.agents/skills/
Windsurf.windsurf/skills/
Agent SDK.claude/skills/
HTTP / API
This release
Versionnot versioned
Publishedno release date on file
PriceFree
Referencenexu-io/ecommerce-image-workflow

Versions

Its author publishes no version number, so there is nothing here to pin to: what you install is the folder as it stands today. Instructions change more often than APIs do — a skill can be rewritten entirely without anything it depends on moving.

v
  • No earlier releases have been published to the marketplace.
Pinning

Nothing to pin to: this skill carries no version number of its own. What you install is what the folder holds on the day you install it.

Reviews

—no reviews yet · no installs yet

Nobody has reviewed this skill yet. The rating is the mean of the reviews written here, so there is none until somebody writes the first.

Who can post

Only accounts that have had the skill installed for fourteen days, so a review is written after living with it rather than after reading it. Publishers may reply once.

Who wrote it

NE
nexu-io

Publishes on mcprush.

0 servers listed28 skills listednot claimed
Profile
Publisher
Servers0