eikona-auctra-visual-router

Routes Auctra content and story-bible assets into Eikona image workflows with review handoff.

Updated Aug 3, 2026
One-click install
npx skills add https://github.com/yeisme/auctra-novel-skills --skill eikona-auctra-visual-router-yeisme
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: eikona-auctra-visual-router
Source: https://github.com/yeisme/auctra-novel-skills/tree/main/auctra-visual/eikona-auctra-visual-router
Command: npx skills add https://github.com/yeisme/auctra-novel-skills --skill eikona-auctra-visual-router-yeisme

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning accepted Auctra content, novel settings, and story-bible entities into images requires coordinating two separate systems: Auctra owns content context, briefs, and review, while Eikona owns prompt execution and provider runs. This Skill defines the routing rules, profiles, and CLI command sequences so visual requests move through brief creation, review acceptance, export, Eikona workflow execution, and candidate handoff without skipping review gates or leaking canon. ## Core Features & Use Cases - Profile-based routing: Maps requests to profiles such as xhs_card_series, wechat_cover, novel_character_sheet, novel_location_map, novel_timeline_board, novel_relationship_graph, novel_world_background, novel_style_pack, and short_video_storyboard. - Brief-to-run pipeline: Creates Auctra visual briefs, accepts reviews, exports briefs for Eikona, then imports, validates, runs, and reviews Eikona workflows with run evidence and feedback. - Prompt collection workflows: Supports multi-candidate prompt file sets under prompts/auctra/<asset-type>/<brief-id>/ with Eikona-managed runbooks, dry runs, and background execution. - Use Case: Given an accepted novel chapter and a character entity, generate a novel_character_sheet brief, export it to Eikona, run the workflow with openai/gpt-5.4-image-2, then import the resulting candidate back into Auctra for acceptance. ## Quick Start Ask the agent to create an Auctra visual brief for a content unit with the novel_character_sheet profile, export it to Eikona, run the workflow, and import the candidate back for review.

Frequently Asked Questions about eikona-auctra-visual-router

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate novel character sheet images from Auctra content?▼

Create a visual brief with the novel_character_sheet profile using auctra visual brief or auctra story bible visual-brief, accept the review, export the brief for Eikona, then import, validate, and run the Eikona workflow. Finished candidates return to Auctra via visual import-handoff.

What image model does the Eikona Auctra workflow use?▼

The default real remote model is openai/gpt-5.4-image-2, with gpt-5.4-image-2 and gpt-image-2 as compatible short aliases. Local validation can use fixture workflows or dry runs without calling a remote provider.

Can I run multiple prompt candidates for one visual brief?▼

Yes. Build a prompt collection under prompts/auctra/<asset-type>/<brief-id>/ with individual prompt files and an Eikona-managed runbook. Each job's prompt, prompt_file, and prompt_ref are mutually exclusive, and every file must trace back to accepted Auctra source refs.

Does this workflow support Scaena production visuals?▼

Only after a current passed preflight from the scaena-subject-asset-readiness check. Without preflight, the router stops and only allows planning, look development, or non-production candidates; production prompts and real generation commands are blocked.

When should I not use Eikona for Auctra image requests?▼

Eikona is the only default image path for Auctra visual requests. Fallback to other tools is allowed only when Eikona is unavailable and the user explicitly approves the fallback.