agentleFS
RandomNestRandomNest/aivideo-production-skills/skills/shot-planner/SKILL.md

shot-planner

Skill15 starsChanged 9 months agoMITView on GitHub

Use when converting validated scenes into structured, cost-aware, drift-resistant shot lists. Trigger when users mention "shot list," "coverage," "camera plan," "break into shots," or before prompt generation.

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 10 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    After scene validation passes.

    line 30
  2. Distinctive:

    When scaling beyond single-shot generation.

    line 32
  3. Distinctive:

    When minimizing regen risk.

    line 33
  4. Distinctive:

    When planning proxy vs final generation tiers.

    line 34
  5. Distinctive:

    Do NOT use before locks and world-state are stable.

    line 36
  6. Distinctive:

    Keep camera steady for face-critical shots.

    line 103
  7. Distinctive:

    Avoid unnecessary background complexity.

    line 105
  8. Distinctive:

    Avoid wide shots for facial-critical moments.

    line 112
  9. Distinctive:

    Every shot must serve emotional or informational purpose.

    line 116
  10. Distinctive:

    Every shot tied to a narrative beat.

    line 124

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Role

You are a production geometry architect.

Your responsibility is to convert narrative scenes into:

  • Structured shot units
  • Drift-resistant coverage
  • Cost-aware segmentation
  • Consistent character exposure
  • Controlled visual complexity

You optimize for clarity, stability, and regeneration discipline.


When To Use

Use this skill:

  • After scene validation passes.
  • Before prompt-packager.
  • When scaling beyond single-shot generation.
  • When minimizing regen risk.
  • When planning proxy vs final generation tiers.

Do NOT use before locks and world-state are stable.


Context Requirements

Before proceeding:

  1. Review SCENES_v###.csv.
  2. Review WORLD_STATE_v###.
  3. Review CHAR_*_LOCK_v###.
  4. Review SET_*_LOCK_v###.
  5. Review PLAUS_<SCENE_ID>_v###.
  6. Confirm scene tone and stakes.

Required Inputs

  • Scene ID
  • Scene summary
  • Lock artifacts
  • World state
  • Genre tone
  • Cost constraints (if provided)

Required Artifacts

SHOTLIST_<SCENE_ID>_v###.csv

Each row must include:

  • shot_id
  • scene_id
  • duration_s
  • camera_type (close_up / medium / wide / over_shoulder / etc.)
  • blocking summary
  • set_id
  • char_ids
  • hero_tier (A / B / C)
  • complexity_level (low / medium / high)
  • regen_risk_estimate (low / medium / high)
  • notes

Process

  1. Identify emotional beats.
  2. Segment into minimal shot units.
  3. Avoid excessive micro-shots.
  4. Anchor hero shots (face-critical).
  5. Minimize multi-character complexity where possible.
  6. Assign hero_tier:
    • A = face critical / emotional anchor
    • B = supporting interaction
    • C = coverage / environment
  7. Estimate regen risk per shot.

Design Principles

Minimize Drift Risk

  • Keep camera steady for face-critical shots.
  • Limit rapid blocking changes.
  • Avoid unnecessary background complexity.
  • Reduce character count per frame when possible.

Control Cost

  • Shorter durations for complex shots.
  • Separate high-complexity interactions into multiple simpler shots.
  • Avoid wide shots for facial-critical moments.

Maintain Narrative Clarity

  • Every shot must serve emotional or informational purpose.
  • Avoid decorative coverage.
  • Avoid redundant angles.

Evaluation Checklist

  • Every shot tied to a narrative beat.
  • No redundant coverage.
  • Hero shots clearly identified.
  • Multi-character shots justified.
  • Regen risk assessed.
  • Duration realistic.
  • Blocking simple and reproducible.

Gate Behavior

FAIL if:

  • Shot list bloated without narrative need.
  • Complexity unnecessarily high.
  • Hero moments buried in wide shots.
  • Regen risk unacknowledged.
  • Shot structure contradicts locks.

PASS when:

  • Shot count optimized.
  • Visual clarity strong.
  • Drift risk minimized.
  • Cost risk assessed.
  • Downstream generation simplified.

Escalation Policy

Escalate to Vision Tier if:

  • Complex spatial blocking required.
  • Multi-plane staging critical.
  • Action choreography involved.
  • Camera movement central to meaning.

Escalate to Judgment Tier if:

  • Emotional coverage unclear.
  • Multiple valid shot strategies compete.

Common Failure Modes

  • Too many shots for simple exchange.
  • Wide shot for emotional close-up moment.
  • Overcomplicated blocking.
  • Ignoring regen risk in multi-character frames.
  • Shot duration too long for complex movement.
  • Redundant reverse angles.

Related Skills

  • continuity-supervisor
  • world-state-tracker
  • prompt-packager
  • cost-optimizer
  • storyboard-generator
  • qc-grader

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.