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.
In this repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- audience-simulatorSkill
- budget-gatesSkill
- character-lockSkill
- character-psychologySkill
- continuity-supervisorSkill
- cost-optimizerSkill
- cultural-checkSkill
- dialogue-consistency-checkSkill
- model-routerSkill
- narrative-red-teamSkill
- plausibility-auditorSkill
- prompt-packagerSkill
- qc-graderSkill
- regen-analyzerSkill
- research-agentSkill
- run-loggerSkill
- script-breakdownSkill
- set-lockSkill
- story-architectSkill
- storyboard-generatorSkill
- voice-lockSkill
- world-state-trackerSkill
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
- Distinctive: line 30
After scene validation passes.
- Distinctive: line 32
When scaling beyond single-shot generation.
- Distinctive: line 33
When minimizing regen risk.
- Distinctive: line 34
When planning proxy vs final generation tiers.
- Distinctive: line 36
Do NOT use before locks and world-state are stable.
- Distinctive: line 103
Keep camera steady for face-critical shots.
- Distinctive: line 105
Avoid unnecessary background complexity.
- Distinctive: line 112
Avoid wide shots for facial-critical moments.
- Distinctive: line 116
Every shot must serve emotional or informational purpose.
- Distinctive: line 124
Every shot tied to a narrative beat.
≈ “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.
Files
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:
- Review SCENES_v###.csv.
- Review WORLD_STATE_v###.
- Review CHAR_*_LOCK_v###.
- Review SET_*_LOCK_v###.
- Review PLAUS_<SCENE_ID>_v###.
- 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
- Identify emotional beats.
- Segment into minimal shot units.
- Avoid excessive micro-shots.
- Anchor hero shots (face-critical).
- Minimize multi-character complexity where possible.
- Assign hero_tier:
- A = face critical / emotional anchor
- B = supporting interaction
- C = coverage / environment
- 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
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
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.

