SKILLEMALL.ai
GitHub ↗ RU Sign in

CC json-prompt-engine

Analyze a reference image (or video frame) and emit a ready-to-paste structured JSON prompt that captures every visual quality needed to reproduce it through `ralphy generate image --prompt-file`. The reverse of the prompt cookbook: image in -> dense JSON prompt out (scene / style / technical / materials / environment / composition / quality blocks; add a `motion` block for image-to-video). Adapts the schema to ralphy's model stack — `gemini-3-pro-image` for multi-ref consistency, `gpt-5.4-image-2` for crisp typography — and folds any matching `@guideline:<slug>` rules into the quality block. USE WHEN the user uploads or points at a reference image and asks for a prompt, "JSON prompt", "image-to-prompt", "reverse-engineer this image", "recreate this image", "describe this in prompt format", "give me the prompt for this style/look", or wants a per-slide JSON prompt for a carousel. DO NOT FIRE for conversational "what is in this image" questions, image editing, or when the user already has a brief and only wants generation (that is the art-director playbook). This runs THROUGH `ralphy generate`, it does not bypass the reference-required gate or the quality gate.

alecs5am/ralphy Agent Skills author: alecs5am 2 files body ≈ 3 701 tokens Open the sourcegithub.com↗ analyzed 11 h ago

Analyze a reference image (or video frame) and emit a ready-to-paste structured JSON prompt that captures every visual quality needed to reproduce it through…

As a process C 54/100 · Has gaps — References files that are not bundled: ../../../.agents/skills/art-director/SKILL.md, ../poster/SKILL.md, ../carousel/SKILL.md

AnalyzerAI and agentstype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
C
82/100
safety, quality, tests
Safety 60%
100
Quality 40%
55
Run on models
none yet
Process rating
C
54/100
Has gaps
References files that are not bundled: ../../../.agents/skills/art-director/SKILL.md, ../poster/SKILL.md, ../carousel/SKILL.md
Tools and files w 18
0
Inputs and preconditions w 11
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

How to improve

  1. Shorten the description to 1024 characters.
  2. The text references files that are not there: add them or drop the references.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 0

✓ No critical or high findings

Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • error description-long description is 1178 chars, limit 1024
  • warning missing-ref reference to a missing file: ../../../.agents/skills/art-director/SKILL.md
  • warning missing-ref reference to a missing file: ../poster/SKILL.md
  • warning missing-ref reference to a missing file: ../carousel/SKILL.md
  • warning missing-ref reference to a missing file: ../fb-creatives/SKILL.md
  • note frontmatter-key unknown frontmatter key "namespace"

Process rating: all ten parameters 54/100

Will not run. References files that are not bundled: ../../../.agents/skills/art-director/SKILL.md, ../poster/SKILL.md, ../carousel/SKILL.md
  • 0Tools and files. 4 referenced file(s) missing: ../../../.agents/skills/art-director/SKILL.md, ../poster/SKILL.md, ../carousel/SKILL.md
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Progress reporting. Says nothing while it works
  • 30Running it twice. 2 mutating operations with no state check
  • 60Result and completion. Output format stated, no completion criterion
  • 65Failures and branches. 3 branches
  • 70When it triggers. States when to use, but not when not to
  • 100Steps. 37 steps
  • 100Consistency. Name and required fields are in place
  • 100Execution cost. Instruction body is 3701 tokens
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 12 top-level sections: this looks like several domains in one skill
  • low The response is described with custom markup (6 tags): a typed call is more reliable

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +4Description does not say when NOT to use the skill (false activations)
  • +3Description length 1178: 120–800 characters recommended
  • +1No license
  • +2Single-language instructions
  • +5Description quotes 7 example trigger phrases
  • +4Structure: 15 headings
  • +3Step-by-step instructions: 37 items
  • +3Output format is stated explicitly
  • +4Has examples (3 code blocks)
  • +4Reference files are cited in the instructions (1 of 1)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 55.