sethmblack/paks-skills/cinematic-grounding
Claude Code and agent skill by sethmblack/paks-skills
- community
Checking that the SKILL.md file is still in the repository.
Transform abstract concepts into vivid, film-like scenes. Ground every abstraction in specific objects, places, sounds, and sensations.
Install
npx vskill@latest install sethmblack/paks-skills/cinematic-groundingSKILL.md
The same SKILL.md was published earlier as sethmblack/paks-skills/cinematic-grounding.
Cinematic Grounding
A methodology for transforming abstract concepts into vivid, film-like scenes. Ground every abstraction in specific objects, places, sounds, and sensations.
When to Use This Skill
- Content feels abstract, theoretical, or disconnected from physical reality
- Marketing copy sounds corporate and forgettable
- Technical writing lacks human connection
- Emotional content tells rather than shows
- User requests: "make this concrete," "ground this," "make this visual," "show don't tell"
Step-by-Step Methodology
Step 1: Identify the Abstraction
Find the abstract concepts hiding in the content:
- Emotions: "loneliness," "hope," "fear," "joy"
- States: "innovation," "transformation," "growth," "decline"
- Qualities: "excellence," "reliability," "trust," "quality"
Ask: What word here could mean anything to anyone? That's your target.
Step 2: Ask the Filmmaker's Question
For each abstraction, ask: "Where would I film this scene?"
- What time of day is it?
- What's the weather?
- What objects are in the frame?
- What sounds are audible?
- What does it smell like?
- Who is there? What are they wearing?
Step 3: Find the Concrete Anchor
Choose ONE specific image that embodies the abstraction:
| Abstraction | Weak Grounding | Cinematic Grounding | |
|---|---|---|---|
| Loneliness | A sad person | A payphone ringing in an empty parking lot at 3 AM, and nobody's answering | |
| Hope | Feeling optimistic | A kid's first bike, training wheels still on, leaning against a sunrise-lit porch | |
| Innovation | New ideas | A garage at midnight, coffee rings on blueprints, the smell of solder | |
| Trust | Reliability | The same mechanic who fixed your dad's car, thirty years of grease under his nails |
Step 4: Add Sensory Texture
Layer in at least two senses beyond sight:
- Sound: What's playing? What's humming? What's silent?
- Smell: Coffee? Rain on asphalt? Old paper?
- Touch: Temperature? Texture? Weight?
- Taste: (When appropriate)
Step 5: Embed the Original Meaning
The final image should carry the original meaning without naming it. The reader should feel the abstraction without being told what to feel.
Test: Remove any direct reference to the abstraction. Does the image still convey it? If yes, you've succeeded.
Examples
Example 1: Corporate Mission Statement
Abstract input: "We are committed to delivering innovative solutions that empower our customers to achieve their goals in a rapidly changing marketplace."
Cinematic grounding: "We're in the business of giving folks a fighting chance—a wrench when the engine's smoking, a map when the road disappears. The world's spinning faster than a carnival ride run by a one-eyed man who lost his glasses, and we figure someone ought to be handing out flashlights."
Example 2: Emotional State
Abstract input: "She felt overwhelmed by the demands of modern life."
Cinematic grounding: "She stood at the kitchen sink at 6 AM, phone buzzing on the counter, coffee going cold, watching the neighbor's dog bark at nothing—and for a moment she couldn't remember if she'd slept or just blinked too long."
Example 3: Technical Concept
Abstract input: "Our platform provides seamless integration capabilities."
Cinematic grounding: "Think of it like the old railroad switching yards—all those different tracks, different gauges, different schedules—and somewhere in the middle, a switchman who knows which lever moves which car where. That's us. We're the switchman."
Key Principles
- Specificity over generality - "Cadillac" not "car." "Thunderbird" not "wine." "3 AM" not "late."
- Scene over statement - Show the world where the feeling lives
- Objects carry meaning - A broken watch says more than "mortality"
- Time and place anchor emotion - "Tuesday morning" hits different than "one day"
- Trust the image - Don't explain what it means; let it resonate
Anti-Patterns to Avoid
- Explaining the metaphor - If you have to say "this represents...," start over
- Generic imagery - Sunsets, hearts, roads (unless made specific)
- Mixed scenes - One grounding image per abstraction; don't blur locations
- Forgetting the body - Everything has temperature, texture, smell
- Sentiment without sensation - Feelings need physical anchors
Integration with Tom Waits Expert
This skill embodies the core Tom Waits technique: finding the concrete in the abstract, the specific in the general, the film scene in the feeling. When using the tom-waits expert, cinematic grounding is applied automatically. This standalone skill allows the technique to be used without the full voice transformation.
Checklist
Before finalizing cinematically grounded content:
- [ ] Every abstraction has a specific, visual anchor
- [ ] At least two senses are engaged beyond sight
- [ ] Time and place are specific (not "one day" or "somewhere")
- [ ] Objects are named specifically (brand, model, condition)
- [ ] The meaning is felt, not explained
- [ ] The image could be filmed—you can see the shot
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingcinematic-grounding/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/sethmblack/paks-skills/cinematic-grounding)