Self-hosted, Self-Improving, Federated Everything App — dev management, agent orchestration, persistent mind, digital twin, second brain, and creative pipeline from a single dashboard over Tailscale to port the user into the OS
JavaScript ai-agentsdashboarddeveloper-toolsdigital-twinquantified-selfsecond-brainself-hostedtailscale
// Mood-board analysis helpers shared by the per-item prompt-from-media flow
// and the board-level "analyze, then compose a poster" flow.
//
// Pure: no React, no fetches. The page persists whatever `moodBoardAnalysisFromResult`
// returns; the style panel uses `boardAnalyzePlan` to decide whether the next
// click runs vision or only composes from prompts already stored on the items.
import { moodBoardItemAnalysisSource } from './moodBoardItemSrc';
import { IMAGE_GEN_MODE } from './imageGenModes';
// Landscape 3:2 — a board poster, not a portrait character sheet. The image
// service is whichever backend the caller resolved; only the frame is fixed.
export const BOARD_POSTER_SIZE = Object.freeze({ width: 1536, height: 1024 });
/**
* The analysis object stored on a media item after one prompt-from-media run.
* Prefers the prompt that matches the item's own type (video items keep the
* video prompt) and falls back to the other target when the model only filled
* one. Returns null when neither prompt is usable.
*/
export function moodBoardAnalysisFromResult(item, result) {
if (!result) return null;
const preferVideo = item?.type === 'video';
const primary = preferVideo ? result.videoPrompt : result.imagePrompt;
const fallback = preferVideo ? result.imagePrompt : result.videoPrompt;
const usedPrimary = primary != null && primary !== '';
const prompt = usedPrimary ? primary : fallback;
if (!prompt) return null;
const negative = usedPrimary
? (preferVideo ? result.videoNegativePrompt : result.imageNegativePrompt)
: (preferVideo ? result.imageNegativePrompt : result.videoNegativePrompt);
return {
prompt,
negativePrompt: negative || null,
rationale: result.rationale || null,
providerId: result.providerId || null,
model: result.model || null,
};
}
/**
* Resolves the primary prompt text for a mood board item.
* Prefers the analyzed prompt from prompt-from-media, then explicit item.prompt,
* then a gallery item's caption prompt. A pin imported from X, Pinterest or
* another page (or a frame cut from one) never offers its caption: that text is
* someone else's, so only an analysis inferred from its pixels is a prompt.
* Same imported-pin rule as the server's `moodBoardItemCaption`. Returns empty string if no prompt exists.
*/
export function moodBoardItemPrompt(item) {
if (item?.analysis?.prompt && typeof item.analysis.prompt === 'string' && item.analysis.prompt.trim()) {
return item.analysis.prompt.trim();
}
if (item?.prompt && typeof item.prompt === 'string' && item.prompt.trim()) {
return item.prompt.trim();
}
const local = typeof item?.mediaKey === 'string' && /^(?:image|video):/.test(item.mediaKey);
const external = typeof item?.source === 'string' && /^(?:https?:|frame:)/i.test(item.source);
if (local && !external && typeof item.caption === 'string' && item.caption.trim()) {
return item.caption.trim();
}
return '';
}
/**
* Returns true if the item has an explicit prompt-from-media analysis saved.
*/
export function isMoodBoardItemAnalyzed(item) {
return Boolean(item?.analysis?.prompt && typeof item.analysis.prompt === 'string' && item.analysis.prompt.trim());
}
/**
* Returns true if the item has been analyzed OR already has a prompt.
*/
export function moodBoardItemHasPrompt(item) {
return isMoodBoardItemAnalyzed(item) || Boolean(moodBoardItemPrompt(item));
}
/**
* Split a board's items into the ones prompt-from-media can still read and
* the ones that already carry a prompt. External URL pins can't be read, so
* they count as skipped rather than pending.
*/
export function boardAnalyzePlan(items) {
const list = Array.isArray(items) ? items : [];
const pending = [];
let analyzed = 0;
let skipped = 0;
for (const item of list) {
if (moodBoardItemHasPrompt(item)) {
analyzed += 1;
continue;
}
if (item?.type !== 'image' && item?.type !== 'video') continue;
if (moodBoardItemAnalysisSource(item)) pending.push(item);
else skipped += 1;
}
return { pending, analyzed, skipped };
}
// Stable key for the style a poster was queued against, so a completion that
// lands after the user composed a different prompt is not pinned.
export function posterStyleKey(style) {
return JSON.stringify({
prompt: typeof style?.prompt === 'string' ? style.prompt.trim() : '',
negative: typeof style?.negativePrompt === 'string' ? style.negativePrompt.trim() : '',
});
}
/**
* Render config for the board poster. `mode` (when set) overrides the resolved
* install default so the poster can run on any enabled image service. The
* override is explicit (`inheritedBackend: false`) — a mood board has no
* universe/music-video tag that would re-resolve the backend server-side.
* Switching off local drops the local model id so it is not sent to a cloud CLI.
*/
export function boardPosterRenderCfg(imageCfg, mode) {
const chosen = mode || imageCfg?.mode;
const local = chosen === IMAGE_GEN_MODE.LOCAL;
return {
...(imageCfg || {}),
mode: chosen,
inheritedBackend: false,
modelId: local ? (imageCfg?.modelId || null) : null,
cloudModel: null,
width: BOARD_POSTER_SIZE.width,
height: BOARD_POSTER_SIZE.height,
};
}
/**
* Generator handoff links for a board item's card: text-to-image and
* text-to-video need a prompt (the analyzed one, else caption); image-to-image
* and image-to-video also need the item to be a local gallery image. Each key
* is null when that handoff doesn't apply. The negative prompt rides along only
* when the analysis saved one.
*/
export function moodBoardItemSendLinks(item) {
const prompt = moodBoardItemPrompt(item);
const negative = typeof item?.analysis?.negativePrompt === 'string' ? item.analysis.negativePrompt.trim() : '';
const source = moodBoardItemAnalysisSource(item);
const imageFile = item?.type === 'image' && source?.kind !== 'video' ? source?.filename : null;
const build = (path, extra = {}) => {
const params = new URLSearchParams();
if (prompt) params.set('prompt', prompt);
if (negative) params.set('negativePrompt', negative);
for (const [k, v] of Object.entries(extra)) params.set(k, v);
return `${path}?${params}`;
};
return {
textToImage: prompt ? build('/media/image') : null,
imageToImage: imageFile ? build('/media/image', { initImageFile: imageFile }) : null,
video: imageFile ? build('/media/video', { sourceImageFile: imageFile }) : (prompt ? build('/media/video') : null),
};
}