Agent skill · Testing & QA

cinematic-shot-direction

Direct cinematic shots and coverage for AI-assisted content production. Use when an agent must translate story intent into framing, shot size, lens and perspective, camera position, movement, blocking, screen direction, lighting-aware coverage, continuity, shot specifications, or diagnose why generated or filmed shots do not cut together.

Calesthio43,316★ · +2,384/wk · 2 repos on radarProfile →
claude-codecodexcopilotcursorMIT
Install
npx skills add calesthio/generative-media-skills --skill cinematic-shot-direction --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 2
SKILL.md size: 30 KB
Bundled scripts: none
Path: skills/production/creative-direction/cinematic-shot-direction/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 112 · +8 this week
Language: Python
Read our review of the source →

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

Review
written from the skill's own SKILL.md · Aug 5, 2026

What it does

Direct cinematic shots and coverage for AI-assisted content production. Use when an agent must translate story intent into framing, shot size, lens and perspective, camera position, movement, blocking, screen direction, lighting-aware coverage, continuity, shot specifications, or diagnose why generated or filmed shots do not cut together.

How it works

Begin with a dramatic job sentence that states what the audience should notice, understand, and feel at a beat. Then identify the shot’s editorial function (Orient, Reveal, Witness, Identify, Compare, Isolate, Transition). Prefer a locked camera if movement has no specific job; movement should be motivated by the script and intended shot.

Build the frame from relationships by selecting shot size based on information, not prestige, and name framing landmarks for precision. Separate distance, height, and angle as independent controls, specify camera position as point of view, and choose lens/perspective in an order: place camera for perspective, choose angle of view, set aperture/focus behavior, then evaluate optical character after geometry works.

Choreograph movement and blocking together by first choosing what moves (subject, camera, both, or neither), and name movement geometrically with start frame, path, motivation, speed, end frame, and newly visible elements. Block with marks and beats, rehearse as a system, and track lighting and key/fill relationships. Preserve screen direction and editorial continuity by defining active axes and recording for each shot who is screen-left/right, gaze target, entry/exit edge, motion direction, axis side changes, props and action state, focus and lighting state, and start/end poses. Use match on action and eyeline, maintain continuity, and justify axis changes if needed.

Design coverage around beats: determine whose information controls the beat, whether geography must remain visible, if decisive events are simultaneous, uninterrupted duration, and editor entry/exit points. A robust minimal set often includes an orienting view, action/speaker view, reaction view, inserts for critical details, and transitions.

Write a generative production shot specification: describe one continuous shot per specification, keeping story context outside visual instructions when not affecting appearance. Use the provided order (SHOT ID / EDITORIAL FUNCTION, Intent, Duration, Subject and setting, Start frame, Lens / focus, Blocking, Camera movement, Lighting interaction, Continuity, Exclusions / tolerances, Review target) and prioritize a single dominant subject action and camera idea. If two moves are essential, state their sequence and cue rather than listing as simultaneous.

When to use it

Use when the task requires translating narrative intent into concrete cinematic direction, diagnosing misaligned shots, or ensuring continuity across generated or filmed coverage.

What it can touch

Describe and specify camera position, movement, shot size, lens, blocking, eyeline, axis direction, lighting states, and continuity conventions for each shot as outlined in the specification structure.

Caveats

License: MIT. The skill emphasizes production heuristics and editorial conventions; it requires adherence to documented editorial functions, shot sizes, and axis rules, and notes when to deliberately cross continuity lines (Production convention). No guarantees of outcomes are stated beyond the procedural guidance.

From the SKILL.md

# Direct cinematic shots as editorial decisions Treat a shot as a controlled change in audience knowledge, attention, and feeling—not as a bundle of cinematic adjectives. Decide what the audience must perceive, choose the camera relationship that makes that perception likely, and specify only the variables needed to preserve it through production and editing. Use three evidence labels throughout t

More from generative-media-skills
All skills →
About this skill
What does the cinematic-shot-direction skill do?

Direct cinematic shots and coverage for AI-assisted content production. Use when an agent must translate story intent into framing, shot size, lens and perspective, camera position, movement, blocking, screen direction, lighting-aware coverage, continuity, shot specifications, or diagnose why generated or filmed shots do not cut together.

How do I install it?

Run `npx skills add calesthio/generative-media-skills --skill cinematic-shot-direction --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From calesthio/generative-media-skills, a repository with 112 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going