Brief Builder

Interviews you until intention, look, camera, cast and deliverables are locked, then writes the one brief every other skill reads.

Creator / AI Mediacreatorai-media
by AgentVoltv1.0.0Published 6d ago1 install
Free to sign up · every skill included with AgentVolt Pro

About this skill

brief-builder (formerly intention-builder)

What it does. Interrogates a filmmaker in numbered rounds, every question carrying a recommended answer, until nothing about the film is silently assumed: intention, ending, look and feel, camera, lens, light, palette, costume, performance, beats, sound, deliverables. Then it checks every answer against the others for contradictions and writes ONE markdown brief (PDF on request) carrying paste-ready cast, location and style tokens.

Why it exists. So you define the style, the look and what you want to say once, and never re-explain it to the next skill or the next session. Every downstream skill reads the brief first.

Generates nothing. No images, no prompts for a model. That starts at reference-builder.

Inputan idea, a vague brief, a reference film, "make it cinematic"
Output<project>-intention.md — the brief; optional PDF
Feedsreference-builder, action-builder, mograph-builder, remotion-builder
Spendsnothing

Files

  • SKILL.md — the interview protocol, the two gate questions, the coherence pass
  • references/question-tree.md — the full question bank by route (film · commercial · explainer …)
  • references/look-library.md — named looks with the camera, light and palette each implies
  • references/format-archetypes.md, handoff-formats.md, coherence-check.md
  • scripts/lint_intention.py — lints a brief for holes and contradictions before handoff
  • assets/ — brief template and PDF styling

Setup

None. Python 3 for the linter only.

SKILL.md preview

SKILL.md
---
name: brief-builder

… (sign up to view the full skill)
Sign up to view, copy, and install the full skill

More video & editing skills

View all Video & Editing skills →