AI & LLM Skill
Agent Eval Suite Langsmith
Production agent eval suite LangSmith dataset curation + Promptfoo assertion framework +…
$15
Prompt Architect
AI image and video prompt builder that guides users from a rough vision to a copy-paste ready…
A structured assistant that takes you from a rough visual idea to a copy-paste ready prompt for AI image and video generation through a clean three-step flow: free-form vision input, prompt generation with gap detection, and a focused iteration loop. It never generates images itself; it builds and refines the prompt, optimized by default for Nano Banana Pro images and Kling video. When inputs conflict or are over-specified, it flags the issue precisely and offers concrete next directions.
$15 one-time
Add to a kit → Prices include 20% VAT. · Forged on real agency work · one-time, no lock-in
- Type Skill
- Category AI & LLM
- Delivery Email · instant
- License One-time
Inside the run · no black box
See the actual work before you buy it.
Describe the image in your own words and skip the twenty-question intake. The skill confirms what is defined, drafts a model-ready prompt, then flags gaps and conflicts before a render burns credits.
- Opens with free vision input: no structured questionnaire, the user describes what they imagine in their own words, optionally with a reference image for subject or style. Twenty upfront questions kill momentum, so step one asks none.
- Summarizes what is actually confirmed under a 'clearly defined' header, listing only what was explicitly stated or can be confidently inferred, never speculating layers into existence.
- Generates a clean, copy-paste ready prompt block optimized for the target model (Nano Banana Pro for images, Kling for video by default), written tight: specific where confirmed, intentionally broad where the user left room.
- Runs gap detection across the seven prompt layers (subject, environment, camera angle, lighting, lens feel, style and era, motion for video) and surfaces only the missing or vague ones as optional suggestions with 1 or 2 concrete example directions each, not open-ended questions.
- Flags input problems in neutral system language before any paid render burns credits: [CONFLICT] when two cues fight each other (cinematic lighting versus natural daylight), [AMBIGUOUS] when a layer can be read two ways, [OVER-SPECIFIED] when detail overload will distort the output.
- Iterates on feedback surgically: diagnoses which single layer caused the complaint, explains the reasoning in 1 or 2 sentences, updates only that layer and returns the full revised prompt. Rewriting from scratch is reserved for a fundamental change of direction.
One power source. 6 lines out.
prompt-architect · core
core active · 6 lines
- ✓ turning a rough visual i…
Turning a rough visual idea into a usable prompt
- ✓ writing image prompts for
Writing image prompts for Nano Banana Pro
- ✓ writing motion prompts for
Writing motion prompts for Kling video
- ✓ diagnosing why a generated
Diagnosing why a generated result felt off
- ✓ resolving conflicting or
Resolving conflicting or over-specified prompt instructions
- ✓ iterating prompts without
Iterating prompts without rewriting from scratch
Yours to keep.
Drag time forward. Watch what stays.
Forever
That's what owning means.
The rented stack
ai writing tool: subscription
expired · access lostanalytics suite: subscription
expired · access lostdesign platform: subscription
expired · access lost(nothing left)
Your forge
-
Go from vague vision to a ready-to-paste prompt fast
license: perpetual -
Spot missing layers before wasting generation credits
license: perpetual -
Fix the exact prompt layer responsible for a bad result
license: perpetual -
Keep prompts tight where confirmed, open where intentional
license: perpetual
subscriptions expire · deeds don't
Everything in the box.
Pick a piece up. Watch it work.
Three-step flow: vision input, generation plus gap detection, iteration
6 parts · one working system · ships instantly by email
This wasn't forged for everyone.
- Not for you if you'd rather rent a tool than own one.
- Not for you if you want someone else to run your stack.
- Not for you if you're happy guessing.
Creators and marketers who want professional AI image and video prompts without learning prompt syntax by trial and error.
then this was forged for you.Works with
Universal by design: these run in any AI. Delivered in the open Agent Skills + MCP format (native in Claude); ChatGPT, Gemini, Cursor and Copilot adapt the same files their own way.
- Claude Native format
- ChatGPT Adapts via open standards
- Gemini Adapts via open standards
- Cursor Adapts via open standards
- Copilot Adapts via open standards
Catch what's on your mind.
the air is clear. nothing between you and the forge.
catch a spark: the forge will answer
-
I use a model other than Nano Banana Pro or Kling, does it still help?
Yes. It optimizes for those two by default, but you can specify another model and the three-step flow plus the prompt layer checklist (subject, environment, camera, lighting, lens, style, motion) carry over. The layers are model-agnostic craft.
-
How does it improve a prompt beyond just rephrasing what I said?
It runs gap detection against the layer checklist, separates confirmed details from inferred ones in a structured summary, and flags CONFLICT, AMBIGUOUS, or OVER-SPECIFIED inputs with concrete next directions. The iteration loop then fixes the exact layer behind a bad result instead of rewriting from scratch.
-
Will it generate the actual image or video for me?
No, it never generates media itself. It builds and refines the copy-paste ready prompt; you take that prompt to your image or video tool and spend your generation credits there.
-
How is it delivered?
By email right after purchase: ready to run, downloaded instantly, no setup wait.
-
One-time or subscription?
A one-time purchase; no subscription or hidden fees. VAT (20%) is included.
-
Can I get a refund?
As a digital product, it can’t be refunded once downloaded. That’s why we show exactly what’s inside and who it’s for, right here.