Skip to content
Back to skills

Storyweaver Brief Intake

ASecurity

Extract a story or Chronicle request into creative brief, constraints, protocol flags, and human questions while preserving the original prompt.

  • 4 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 9, 2026
ai-agents

Security analysis

A100/100

Scanned October 9, 2026

npx -y skills add jgwill/miadi-orchestration-kit --skill storyweaver-brief-intake --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Storyweaver Brief Intake?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Storyweaver Brief Intake
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jgwill-storyweaver-brief-intake/badge)](https://www.skillsdirectory.com/skills/jgwill-storyweaver-brief-intake)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: storyweaver-brief-intake
description: "Extract a story or Chronicle request into creative brief, constraints, protocol flags, and human questions while preserving the original prompt."
---

# Storyweaver Brief Intake

Use this skill when a user gives a new premise, mission, typed starter, session source, or story request.

## Coordinates

- Primary agent: Creative Brief Archaeologist
- Supporting agent: Cultural Protocol Steward, if sensitive material appears
- Routing agent: Storyweaver Orchestration Architect

## Required Reading Order

1. Raw user prompt or session instruction
2. `inputs/original-prompt.md`, if resuming
3. Existing notes, drafts, transcripts, or issue context
4. `state.md`
5. `protocol-notes.md`, if present

## Workflow

1. Preserve the original prompt in `inputs/original-prompt.md`.
2. Extract premise, deliverable, audience, genre or form, tone, point of view, tense, length, and language.
3. Separate operational instructions from prose content.
4. Identify source expectations, citations, canon notes, privacy boundaries, and consent-sensitive material.
5. Write contradictions, unknowns, and optional human questions.
6. Update `state.md` with route and next recommended skill.

## Artefacts Read

- User prompt or `inputs/original-prompt.md`
- Existing drafts, session records, or research notes
- `state.md`

## Artefacts Written

- `inputs/original-prompt.md`
- `creative-brief.md`
- `constraints.md`
- `questions-for-human.md`, only when needed
- `protocol-notes.md`, when relevant

## Brief Sections

Include:

- Desired outcome
- Current reality
- Natural progression
- Audience layers: academic, engineering, narrative
- Story or Chronicle premise
- Output form and deliverables
- Voice, tone, point of view, style, and length
- Content boundaries
- Sources and provenance expectations
- Operational instructions excluded from prose

## Pause Conditions

- Missing choice changes the story direction.
- Source use or personal material requires human consent.
- Operational instruction conflicts with the requested creative form.

## Acceptance Checklist

- [ ] Original prompt is preserved.
- [ ] Creative brief and operational instructions are separated.
- [ ] Sensitive material has a protocol route.
- [ ] Next skill is recorded in `state.md`.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.