@kolbo/mcp 1.38.0 → 1.40.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@kolbo/mcp",
-  "version": "1.38.0",
+  "version": "1.40.0",
   "description": "Kolbo AI MCP Server - Generate images, videos, music, speech, and sound effects from Claude Code",
   "main": "src/index.js",
   "bin": {

package/skill/GENERATED.md

@@ -1,7 +1,7 @@
 # AUTO-GENERATED — do not edit
 
 This skill/ tree is mirrored from kolbo-code (the single source of truth)
-by .github/workflows/sync-skill-to-plugin.yml — synced from kolbo-code@82f9072.
+by .github/workflows/sync-skill-to-plugin.yml — synced from kolbo-code@d70e982.
 
 It is the skill that 'npx @kolbo/mcp install' deploys into the user's agent.
 To change it, edit packages/opencode/skills/kolbo/ in kolbo-code and push;

package/skill/SKILL.md

@@ -1,5 +1,5 @@
 ---
-version: 0.7.1
+version: 0.7.2
 name: kolbo
 description: |
   Generate, edit, or analyze creative media via the Kolbo AI MCP server:
@@ -275,9 +275,6 @@ Chat renders markdown natively. `![alt](url)` = inline image. `[label](url)` = l
 
 Avoid bare URL dumps and HTML `<table>` grids — canvas already provides a gallery.
 
-- **3D models (`generate_3d`)**: NEVER paste the raw GLB/FBX/OBJ/USDZ file URLs as a link list. The result renders as a card with a live 3D preview + Download buttons ("View in Canvas"). Just say the model is ready and describe it — the user grabs formats from the card.
-- **NEVER surface a raw provider domain** (e.g. `*.fal.media`, `replicate.delivery`) in chat or `.kolbo/production.md`. All Kolbo outputs are Kolbo-hosted; if you ever see a provider URL in a result, treat it as a bug — report it, don't echo it.
-
 **After `generate_creative_director` completes** — share results as individual URLs, one per scene. Do NOT create an HTML grid artifact.
 
 **Always** record every URL in `.kolbo/production.md` — see `references/workflows/production-log.md`.

package/skill/VERSION

@@ -1 +1 @@
-0.7.1
+0.7.2

package/src/apps/html.js

@@ -27,6 +27,49 @@ ${body}
 // ---- shared widget runtime ----
 var KOLBO_LOGO_FALLBACK = ${JSON.stringify(KOLBO_LOGO_SVG)};
 var KOLBO_LOGO = ${JSON.stringify(KOLBO_LOGO_IMG)};
+// ---- shared inline SVG icon set (currentColor, 1em; replaces emoji so glyphs
+// render identically in every host iframe instead of tofu boxes) ----
+function _svg(inner, o) {
+  o = o || {};
+  var f = o.fill ? o.fill : 'none';
+  var s = o.fill ? 'none' : 'currentColor';
+  return '<svg class="k-ic" viewBox="0 0 24 24" width="1em" height="1em" fill="' + f + '" stroke="' + s +
+    '" stroke-width="' + (o.w || 2) + '" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true">' + inner + '</svg>';
+}
+var ICONS = {
+  upload: _svg('<path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><path d="M17 8l-5-5-5 5"/><path d="M12 3v12"/>'),
+  download: _svg('<path d="M21 15v4a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2v-4"/><path d="M7 10l5 5 5-5"/><path d="M12 15V3"/>'),
+  play: _svg('<path d="M8 5v14l11-7z"/>', { fill: 'currentColor', w: 1 }),
+  pause: _svg('<path d="M6 4h4v16H6zM14 4h4v16h-4z"/>', { fill: 'currentColor', w: 1 }),
+  check: _svg('<path d="M20 6 9 17l-5-5"/>', { w: 2.4 }),
+  x: _svg('<path d="M18 6 6 18M6 6l12 12"/>', { w: 2.4 }),
+  warn: _svg('<path d="M10.29 3.86 1.82 18a2 2 0 0 0 1.71 3h16.94a2 2 0 0 0 1.71-3L13.71 3.86a2 2 0 0 0-3.42 0z"/><path d="M12 9v4"/><path d="M12 17h.01"/>'),
+  retry: _svg('<path d="M21 2v6h-6"/><path d="M3 12a9 9 0 0 1 15-6.7L21 8"/><path d="M3 22v-6h6"/><path d="M21 12a9 9 0 0 1-15 6.7L3 16"/>'),
+  edit: _svg('<path d="M12 20h9"/><path d="M16.5 3.5a2.12 2.12 0 0 1 3 3L7 19l-4 1 1-4z"/>'),
+  open: _svg('<path d="M18 13v6a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h6"/><path d="M15 3h6v6"/><path d="M10 14 21 3"/>'),
+  arrowRight: _svg('<path d="M5 12h14"/><path d="M12 5l7 7-7 7"/>'),
+  sparkle: _svg('<path d="M12 3l1.9 5.1L19 10l-5.1 1.9L12 17l-1.9-5.1L5 10z"/>', { fill: 'currentColor', w: 1 }),
+  clock: _svg('<circle cx="12" cy="12" r="9"/><path d="M12 7v5l3 2"/>'),
+  sound: _svg('<path d="M11 5 6 9H2v6h4l5 4z"/><path d="M15.5 8.5a5 5 0 0 1 0 7"/><path d="M19 5a9 9 0 0 1 0 14"/>'),
+  mic: _svg('<path d="M12 2a3 3 0 0 0-3 3v6a3 3 0 0 0 6 0V5a3 3 0 0 0-3-3z"/><path d="M19 10a7 7 0 0 1-14 0"/><path d="M12 19v3"/>'),
+  image: _svg('<rect x="3" y="3" width="18" height="18" rx="2"/><circle cx="8.5" cy="8.5" r="1.5"/><path d="M21 15l-5-5L5 21"/>'),
+  video: _svg('<path d="M23 7l-7 5 7 5z"/><rect x="1" y="5" width="15" height="14" rx="2"/>'),
+  audio: _svg('<path d="M9 18V5l12-2v13"/><circle cx="6" cy="18" r="3"/><circle cx="18" cy="16" r="3"/>'),
+  document: _svg('<path d="M14 2H6a2 2 0 0 0-2 2v16a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2V8z"/><path d="M14 2v6h6"/><path d="M8 13h8"/><path d="M8 17h8"/>'),
+  cube: _svg('<path d="M21 8l-9-5-9 5 9 5z"/><path d="M3 8v8l9 5 9-5V8"/><path d="M12 13v8"/>'),
+  file: _svg('<path d="M14 2H6a2 2 0 0 0-2 2v16a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2V8z"/><path d="M14 2v6h6"/>')
+};
+// Media-kind → icon (accepts model kind / media_type strings).
+function kindIcon(kind) {
+  switch (String(kind || '').toLowerCase()) {
+    case 'image': return ICONS.image;
+    case 'video': return ICONS.video;
+    case 'audio': case 'music': case 'sound': case 'speech': return ICONS.audio;
+    case 'document': case 'doc': return ICONS.document;
+    case '3d': case 'three_d': case 'model': return ICONS.cube;
+    default: return ICONS.file;
+  }
+}
 function esc(s) {
   return String(s == null ? '' : s).replace(/[&<>"']/g, function (c) {
     return { '&': '&amp;', '<': '&lt;', '>': '&gt;', '"': '&quot;', "'": '&#39;' }[c];

package/src/apps/theme.js

@@ -222,6 +222,8 @@ html.k-fullscreen .k-actions { flex: none; padding-top: 8px; }
 .k-btn.ghost { background: transparent; box-shadow: none; border-color: transparent; color: var(--text-muted); }
 .k-btn.ghost:hover { color: var(--text); background: var(--surface-2); }
 .k-btn svg { width: 13px; height: 13px; }
+/* Inline SVG icons (replace emoji) — align with text, inherit color, never shrink. */
+.k-ic { width: 1em; height: 1em; vertical-align: -0.14em; flex: none; }
 .k-btn:disabled { opacity: 0.5; cursor: default; }
 
 /* ---- Inline prompt input (Animate / Edit flows) ---- */

package/src/apps/widgets/generation.js

@@ -38,7 +38,7 @@ const BODY = `
     <div class="k-prompt-row" id="prompt-row">
       <input class="k-input" id="action-input" placeholder="">
       <button class="k-btn primary" id="action-send">Send</button>
-      <button class="k-btn ghost" id="action-cancel">✕</button>
+      <button class="k-btn ghost" id="action-cancel" aria-label="Cancel"></button>
     </div>
     <div class="k-actions" id="actions"></div>
   </div>
@@ -55,6 +55,7 @@ var selected = 0;          // selected result index
 var pollTimer = null;
 
 el('logo').innerHTML = KOLBO_LOGO + '<span>Kolbo</span>';
+el('action-cancel').innerHTML = ICONS.x;
 el('kolbo-link').onclick = function (e) { e.preventDefault(); window.kolbo.openLink('https://app.kolbo.ai'); };
 
 var TOOL_TITLES = {
@@ -85,11 +86,11 @@ function renderChips(sc) {
   var h = modelChipHTML(sc.model, sc.model_icon);
   var s = sc.settings || {};
   if (sc.kind) h += chip(iconFor(sc.kind) + ' ' + sc.kind);
-  if (s.duration) h += chip('⏱ ' + fmtDur(s.duration));
+  if (s.duration) h += chip(ICONS.clock + ' ' + fmtDur(s.duration));
   if (s.resolution) h += chip(esc(s.resolution));
   if (s.aspect_ratio) h += chip(esc(s.aspect_ratio));
-  if (s.audio) h += chip('🔊 audio');
-  if (s.voice) h += chip('🎤 ' + esc(s.voice));
+  if (s.audio) h += chip(ICONS.sound + ' audio');
+  if (s.voice) h += chip(ICONS.mic + ' ' + esc(s.voice));
   if (s.mode) h += chip(esc(s.mode));
   if (sc.count > 1) h += chip('×' + sc.count);
   if (sc.reference_image) h += '<img class="k-ref-thumb" src="' + esc(sc.reference_image) + '" alt="" title="Reference image" onerror="this.style.display=\\'none\\'">';
@@ -97,7 +98,13 @@ function renderChips(sc) {
 }
 function chip(inner) { return '<span class="k-chip">' + inner + '</span>'; }
 function iconFor(kind) {
-  return { image: '🖼', video: '🎬', audio: '🎵', '3d': '🧊', scenes: '🎞' }[kind] || '✨';
+  switch (kind) {
+    case 'image': return ICONS.image;
+    case 'video': case 'scenes': return ICONS.video;
+    case 'audio': return ICONS.audio;
+    case '3d': return ICONS.cube;
+    default: return ICONS.sparkle;
+  }
 }
 
 /* ---------- generating ---------- */
@@ -247,7 +254,7 @@ function renderLinks(urls) {
 // Small hover download button attached to a media cell (per-item downloads —
 // batch grids and CD scenes have no single "current" url for the action row).
 function dlBtnHTML(u) {
-  return '<button class="k-dl" data-dl="' + esc(u) + '" title="Download" aria-label="Download">⬇</button>';
+  return '<button class="k-dl" data-dl="' + esc(u) + '" title="Download" aria-label="Download">' + ICONS.download + '</button>';
 }
 function wireDlButtons(root) {
   Array.prototype.forEach.call((root || document).querySelectorAll('.k-dl[data-dl]'), function (b) {
@@ -280,7 +287,7 @@ function renderScenes(sc) {
     : '<img id="scene-main" src="' + esc(it.url) + '" alt="" style="cursor:zoom-in">';
   var thumbs = '<div class="k-thumbs">' + items.map(function (t, i) {
     var inner = t.type === 'video'
-      ? '<span style="display:flex;align-items:center;justify-content:center;width:100%;height:100%;background:rgba(255,255,255,0.06);font-size:14px">▶</span>'
+      ? '<span style="display:flex;align-items:center;justify-content:center;width:100%;height:100%;background:rgba(255,255,255,0.06);font-size:22px">' + ICONS.play + '</span>'
       : '<img src="' + esc(t.url) + '" alt="" loading="lazy">';
     return '<div class="k-thumb' + (i === selected ? ' active' : '') + '" data-i="' + i + '" title="' + esc(t.label) + '">' + inner + '</div>';
   }).join('') + '</div>';
@@ -324,8 +331,8 @@ function renderError(msg) {
   clearTimeout(pollTimer);
 
   setPhaseChip('Failed', false);
-  el('stage').innerHTML = '<div class="k-error">⚠ ' + esc(msg) + '</div>';
-  el('actions').innerHTML = '<button class="k-btn" id="retry-btn">↻ Try Again</button>';
+  el('stage').innerHTML = '<div class="k-error">' + ICONS.warn + ' ' + esc(msg) + '</div>';
+  el('actions').innerHTML = '<button class="k-btn" id="retry-btn">' + ICONS.retry + ' Try Again</button>';
   el('retry-btn').onclick = function () {
     var what = (state && TOOL_TITLES[state.tool]) || 'generation';
     window.kolbo.sendMessage('Please retry that ' + what.toLowerCase() + ' — it failed with: ' + msg);
@@ -355,7 +362,7 @@ function applyFullscreen(on, exitHandler) {
   var c = el('phase-chip');
   if (on) {
     c.style.display = '';
-    c.innerHTML = '✕ ' + esc('Exit');
+    c.innerHTML = ICONS.x + ' ' + esc('Exit');
     c.style.cursor = 'pointer';
     c.onclick = exitHandler || toggleFullscreen;
   } else {
@@ -382,18 +389,18 @@ function renderActions(sc) {
   var a = [];
   var hasSingleUrl = !!(sc.urls && sc.urls.length);
   if (sc.kind === 'image') {
-    a.push('<button class="k-btn primary" id="btn-animate">🎬 Animate</button>');
-    a.push('<button class="k-btn" id="btn-edit">✏️ Edit</button>');
+    a.push('<button class="k-btn primary" id="btn-animate">' + ICONS.video + ' Animate</button>');
+    a.push('<button class="k-btn" id="btn-edit">' + ICONS.edit + ' Edit</button>');
   }
   if (sc.kind === 'video') {
-    a.push('<button class="k-btn primary" id="btn-download">⬇ Download</button>');
+    a.push('<button class="k-btn primary" id="btn-download">' + ICONS.download + ' Download</button>');
   } else if (hasSingleUrl) {
     // Scenes (Creative Director) have no single "current" url — per-item hover
     // download buttons cover them instead.
-    a.push('<button class="k-btn" id="btn-download">⬇ Download</button>');
+    a.push('<button class="k-btn" id="btn-download">' + ICONS.download + ' Download</button>');
   }
-  a.push('<button class="k-btn" id="btn-recreate">↻ Recreate</button>');
-  a.push('<button class="k-btn ghost" id="btn-open">Open in Kolbo ↗</button>');
+  a.push('<button class="k-btn" id="btn-recreate">' + ICONS.retry + ' Recreate</button>');
+  a.push('<button class="k-btn ghost" id="btn-open">Open in Kolbo ' + ICONS.open + '</button>');
   el('actions').innerHTML = a.join('');
 
   bind('btn-download', function () { window.kolbo.openLink(downloadUrl(currentUrl())); });
@@ -405,7 +412,7 @@ function renderActions(sc) {
       '\\n(from the ' + (TOOL_TITLES[state.tool] || 'generation') + ' widget)');
   });
   bind('btn-animate', function () {
-    openPromptRow('Describe the motion (optional — Smart Select picks the best video model)…', function (text) {
+    openPromptRow('Describe the motion (optional)…', function (text) {
       window.kolbo.sendMessage('Animate this image into a short video' +
         '\\n🎬 Reference image: ' + currentUrl() +
         '\\nModel: smart select — pick the best image-to-video model' +

package/src/apps/widgets/mediaGrid.js

@@ -65,8 +65,8 @@ function cellHTML(item, i) {
   var idx = state.items.indexOf(item);
   var media = item.thumbnail
     ? '<img src="' + esc(item.thumbnail) + '" loading="lazy" alt="">'
-    : '<div style="display:flex;align-items:center;justify-content:center;height:100%;color:var(--text-faint);font-size:20px">' +
-      ({ video: '🎬', '3d': '🧊' }[item.media_type] || '🖼') + '</div>';
+    : '<div style="display:flex;align-items:center;justify-content:center;height:100%;color:var(--text-faint);font-size:22px">' +
+      kindIcon(item.media_type) + '</div>';
   return '<div class="k-cell" data-i="' + idx + '">' +
     '<div class="k-cell-media">' + media + '</div>' +
     '<div class="k-cell-label">' + esc(item.title || '') + '</div>' +
@@ -81,7 +81,7 @@ function audioRowHTML(item) {
     '<div class="k-audio-meta"><div class="k-audio-title">' + esc(item.title || '') + '</div>' +
     '<div class="k-audio-sub">' + esc(item.subtitle || '') + '</div></div>' +
     (item.preview_audio || item.url
-      ? '<button class="k-play" data-play="' + esc(item.preview_audio || item.url) + '">▶</button>' : '') +
+      ? '<button class="k-play" data-play="' + esc(item.preview_audio || item.url) + '">' + ICONS.play + '</button>' : '') +
     '<button class="k-btn" data-use="' + idx + '">Use</button></div>';
 }
 
@@ -96,13 +96,13 @@ function wire() {
     b.onclick = function (e) {
       e.stopPropagation();
       var url = b.getAttribute('data-play');
-      if (playing && playing.src === url && !playing.paused) { playing.pause(); b.textContent = '▶'; return; }
+      if (playing && playing.src === url && !playing.paused) { playing.pause(); b.innerHTML = ICONS.play; return; }
       if (playing) playing.pause();
-      Array.prototype.forEach.call(document.querySelectorAll('[data-play]'), function (x) { x.textContent = '▶'; });
+      Array.prototype.forEach.call(document.querySelectorAll('[data-play]'), function (x) { x.innerHTML = ICONS.play; });
       playing = new Audio(url);
       playing.play();
-      b.textContent = '⏸';
-      playing.onended = function () { b.textContent = '▶'; };
+      b.innerHTML = ICONS.pause;
+      playing.onended = function () { b.innerHTML = ICONS.play; };
     };
   });
 }

package/src/apps/widgets/transcript.js

@@ -54,7 +54,7 @@ function boot(sc) {
   }
   if (sc.phase === 'failed') {
     el('phase-chip').style.display = 'none';
-    el('stage').innerHTML = '<div class="k-error">⚠ ' + esc(sc.error || 'Transcription failed') + '</div>';
+    el('stage').innerHTML = '<div class="k-error">' + ICONS.warn + ' ' + esc(sc.error || 'Transcription failed') + '</div>';
     return;
   }
   el('phase-chip').style.display = '';
@@ -67,9 +67,9 @@ function boot(sc) {
     'background:var(--surface);border:1px solid var(--border);font-size:12.5px;color:var(--text-muted);white-space:pre-wrap">' +
     esc(sc.text || '(empty transcript)') + '</div>';
   var a = [];
-  if (sc.srt_url) a.push('<button class="k-btn primary" data-url="' + esc(sc.srt_url) + '">⬇ SRT</button>');
-  if (sc.word_by_word_srt_url) a.push('<button class="k-btn" data-url="' + esc(sc.word_by_word_srt_url) + '">⬇ Word-by-word SRT</button>');
-  if (sc.txt_url) a.push('<button class="k-btn" data-url="' + esc(sc.txt_url) + '">⬇ TXT</button>');
+  if (sc.srt_url) a.push('<button class="k-btn primary" data-url="' + esc(sc.srt_url) + '">' + ICONS.download + ' SRT</button>');
+  if (sc.word_by_word_srt_url) a.push('<button class="k-btn" data-url="' + esc(sc.word_by_word_srt_url) + '">' + ICONS.download + ' Word-by-word SRT</button>');
+  if (sc.txt_url) a.push('<button class="k-btn" data-url="' + esc(sc.txt_url) + '">' + ICONS.download + ' TXT</button>');
   a.push('<button class="k-btn ghost" id="btn-copy">Copy text</button>');
   el('actions').innerHTML = a.join('');
   Array.prototype.forEach.call(el('actions').querySelectorAll('[data-url]'), function (b) {
@@ -77,7 +77,7 @@ function boot(sc) {
   });
   var copyBtn = el('btn-copy');
   if (copyBtn) copyBtn.onclick = function () {
-    try { navigator.clipboard.writeText(state.text || ''); copyBtn.textContent = 'Copied ✓'; } catch (e) {}
+    try { navigator.clipboard.writeText(state.text || ''); copyBtn.innerHTML = 'Copied ' + ICONS.check; } catch (e) {}
   };
   window.kolbo.notifySize();
 }
@@ -107,7 +107,7 @@ window.kolbo.onToolResult(function (result) {
   if (result.isError || /error|failed/i.test(txt)) {
     el('phase-chip').style.display = 'none';
     el('stage').classList.remove('k-empty');
-    el('stage').innerHTML = '<div class="k-error">⚠ ' + esc((txt || 'Transcription failed').slice(0, 300)) + '</div>';
+    el('stage').innerHTML = '<div class="k-error">' + ICONS.warn + ' ' + esc((txt || 'Transcription failed').slice(0, 300)) + '</div>';
     window.kolbo.notifySize();
     return;
   }

package/src/apps/widgets/upload.js

@@ -33,7 +33,7 @@ const BODY = `
   </div>
   <div class="k-body">
     <div id="drop" style="border:1.5px dashed var(--border);border-radius:12px;padding:26px 16px;text-align:center;cursor:pointer;transition:border-color .15s,background .15s">
-      <div style="font-size:22px;margin-bottom:6px">⬆</div>
+      <div id="drop-icon" style="font-size:26px;line-height:1;margin-bottom:8px;color:var(--text-muted)"></div>
       <div style="font-size:13px;font-weight:600">Click or drop files here</div>
       <div id="accept-hint" style="font-size:11.5px;color:var(--text-muted);margin-top:4px"></div>
     </div>
@@ -51,6 +51,7 @@ const BODY = `
 
 const SCRIPT = `
 el('logo').innerHTML = KOLBO_LOGO + '<span>Kolbo</span>';
+el('drop-icon').innerHTML = ICONS.upload;
 el('kolbo-link').onclick = function (e) { e.preventDefault(); window.kolbo.openLink('https://app.kolbo.ai/media-library'); };
 
 var state = null;
@@ -61,10 +62,10 @@ var active = 0;
 var sent = false;
 
 var KINDS = {
-  image: { exts: ['jpg','jpeg','png','webp','gif','heic','heif','avif','bmp','tif','tiff'], icon: '🖼' },
-  video: { exts: ['mp4','mov','webm','m4v','mkv','avi'], icon: '🎬' },
-  audio: { exts: ['mp3','wav','m4a','aac','ogg','flac'], icon: '🎵' },
-  document: { exts: ['pdf','txt','md','csv','json','docx','xlsx','pptx','doc','xls'], icon: '📄' }
+  image: { exts: ['jpg','jpeg','png','webp','gif','heic','heif','avif','bmp','tif','tiff'], icon: ICONS.image },
+  video: { exts: ['mp4','mov','webm','m4v','mkv','avi'], icon: ICONS.video },
+  audio: { exts: ['mp3','wav','m4a','aac','ogg','flac'], icon: ICONS.audio },
+  document: { exts: ['pdf','txt','md','csv','json','docx','xlsx','pptx','doc','xls'], icon: ICONS.document }
 };
 
 function classify(name) {
@@ -98,7 +99,7 @@ function showExpired() {
   el('drop').style.pointerEvents = 'none';
   el('drop').style.opacity = '0.5';
   el('notice').style.display = '';
-  el('notice').innerHTML = '⏱ This upload window expired. Ask Claude to open a new upload widget.';
+  el('notice').innerHTML = ICONS.clock + ' This upload window expired. Ask Claude to open a new upload widget.';
   window.kolbo.notifySize();
 }
 
@@ -190,12 +191,12 @@ function upload(it) {
 
 // ---- rendering ----
 function rowHtml(it) {
-  var icon = it.kind && KINDS[it.kind] ? KINDS[it.kind].icon : '📎';
+  var icon = it.kind && KINDS[it.kind] ? KINDS[it.kind].icon : ICONS.file;
   var right = '';
   if (it.status === 'queued') right = '<span style="color:var(--text-muted)">queued</span>';
   else if (it.status === 'uploading') right = '<span style="color:var(--text-muted)">' + it.pct + '%</span>';
-  else if (it.status === 'done') right = '<span style="color:#4ade80">✓ uploaded</span>';
-  else right = '<span class="k-error" style="padding:0;border:0;background:none">✕ ' + esc(it.err || 'failed') + '</span> <a href="#" data-retry="' + it.id + '" style="font-size:11px">retry</a>';
+  else if (it.status === 'done') right = '<span style="color:#4ade80">' + ICONS.check + ' uploaded</span>';
+  else right = '<span class="k-error" style="padding:0;border:0;background:none">' + ICONS.x + ' ' + esc(it.err || 'failed') + '</span> <a href="#" data-retry="' + it.id + '" style="font-size:11px">retry</a>';
   var bar = it.status === 'uploading'
     ? '<div style="height:3px;border-radius:2px;background:var(--surface);margin-top:5px;overflow:hidden"><div id="bar-' + it.id + '" style="height:100%;width:' + it.pct + '%;background:var(--accent,#7c6cff);transition:width .2s"></div></div>'
     : '';
@@ -241,7 +242,7 @@ function render() {
       sent = true;
       var lines = done.map(function (i, idx) { return (idx + 1) + '. ' + i.file.name + ' (' + i.kind + '): ' + i.url; });
       window.kolbo.sendMessage('I uploaded ' + done.length + ' file(s) to my Kolbo media library:\\n' + lines.join('\\n') + '\\nContinue with these files.');
-      el('actions').innerHTML = '<span style="font-size:12px;color:var(--text-muted)">✓ Sent to Claude — continuing…</span>';
+      el('actions').innerHTML = '<span style="font-size:12px;color:var(--text-muted)">' + ICONS.check + ' Sent to Claude — continuing…</span>';
       window.kolbo.notifySize();
     };
     el('btn-more').onclick = function () { el('picker').click(); };

package/src/index.js

@@ -119,7 +119,9 @@ function createServer(opts = {}) {
       '5. Written deliverables (plans, briefs, scripts, research summaries) can live in Kolbo too: author them as AI Docs with `create_doc` (project-scoped, editable in the app, shareable via `share_doc`).',
       '6. DIRECTOR / BATCH JOBS: generate_creative_director runs its scenes (image OR video) in parallel and only reports state="completed" once EVERY scene is terminal. Video batches can take many minutes. If the tool returns `_timed_out:true`, the batch is STILL RUNNING on the server — call `get_creative_director_status` with the returned generation_id and keep checking until state="completed" to collect all scene outputs. NEVER conclude a Director run failed and fall back to plain generate_image/generate_video without first checking status — doing so wastes the user\'s credits by paying twice. If scenes already carry image_urls/video_urls, they are done; do not regenerate.',
       '7. SESSION CONTINUITY: keep one workflow in ONE session. chat_send_message and the generation tools return a `session_id` — for follow-ups, refinements, retries, or additional steps on the SAME task/theme, pass that same `session_id` back on the next call instead of starting fresh. Only OMIT session_id (start a new session) when the user genuinely switches to an unrelated task. Do not open a new conversation/session for every message of the same workflow — it fragments the user\'s history and loses context.',
-      '8. LOCAL FILES / CHAT ATTACHMENTS: remote MCP tools CANNOT read files the user attached to the chat. When a claude.ai (browser/mobile) user has a local image/video/audio/document to use as a generation input, IMMEDIATELY call `media_upload_widget` — an upload card appears in the chat, they upload, and stable Kolbo CDN URLs come back in a follow-up message. Never ask them to re-attach the file in chat and never invent a URL. On Claude Desktop/Code with filesystem access, use `upload_media` with the absolute local path instead.'
+      '8. LOCAL FILES / CHAT ATTACHMENTS: remote MCP tools CANNOT read files the user attached to the chat. When a claude.ai (browser/mobile) user has a local image/video/audio/document to use as a generation input, IMMEDIATELY call `media_upload_widget` — an upload card appears in the chat, they upload, and stable Kolbo CDN URLs come back in a follow-up message. Never ask them to re-attach the file in chat and never invent a URL. On Claude Desktop/Code with filesystem access, use `upload_media` with the absolute local path instead.',
+      '9. MODEL SELECTION: ALWAYS pass a specific `model` on every generation tool — do NOT omit it. Omitting falls back to "Smart Select" auto-routing, which we deliberately avoid because it hides the model choice from the user and often picks a generic default. Choose the model that best fits the task and the user\'s intent (quality, speed, style, capability). If you are unsure which model to use for a given type, call `list_models` with the matching `type` and pick the recommended/flagship one, then pass its `identifier`. Only use Smart Select (omit `model`) if the user EXPLICITLY asks you to auto-pick.',
+      '10. IMAGE EDITING: for ANY prompt-driven / content edit of an existing image — "make it night", changing scene/lighting/colors, adding/removing/replacing objects, restyling — use `generate_image_edit` (it runs on strong dedicated editing models, same as image generation). Do NOT use `edit_image` for content edits — `edit_image` is ONLY for mechanical enhancements (upscale, reframe, remove-background, skin retouch). Its `magic_edit` operation is deprecated in favor of `generate_image_edit`.'
     ].join('\n')
   });
 

package/src/tools/generate.js

@@ -52,7 +52,7 @@ function registerGenerateTools(server, client, options = {}) {
     'Generate image(s) from a text prompt using Kolbo AI. Supports Visual DNA profiles (for character/style/product consistency), moodboards (for style direction), reference images (for composition guidance), batch generation (num_images), and web-search grounding. For EDITING an existing image, use generate_image_edit instead. For a coordinated multi-scene set (storyboard, ad campaign), use generate_creative_director. Returns the final image URL(s) when complete.',
     {
       prompt: z.string().describe('Text description of the image to generate'),
-      model: z.string().optional().describe('Model identifier. Use list_models type="text_to_img" to see options. Omit for Smart Select.'),
+      model: z.string().optional().describe('Model identifier — REQUIRED in practice: pick a specific model, do NOT omit (omitting = Smart Select auto-pick, which we avoid). Strong current defaults: "nano-banana-2" (versatile, text rendering, multilingual) or "gpt-image-2" (photoreal, infographics). Call list_models type="text_to_img" to see all options and pick per the user\'s intent.'),
       aspect_ratio: z.string().optional().describe('Aspect ratio (e.g., "1:1", "16:9", "9:16"). Must be a value present in the model\'s `supported_aspect_ratios` from list_models — pass an unsupported value and the API rejects. Default: "1:1"'),
       enhance_prompt: z.boolean().optional().describe('Enhance the prompt for better results. Default: true'),
       num_images: z.number().optional().describe('Number of images to generate in one call. Default: 1. Note: some models (Midjourney etc.) have a fixed `images_per_request` and ignore this — check list_models.'),
@@ -92,7 +92,7 @@ function registerGenerateTools(server, client, options = {}) {
             urls: result.result.urls,
             model: result.result.model,
             prompt_used: result.result.prompt_used,
-            _followup_hint: 'If the user asks to edit/change/modify this image next, pass urls[0] to generate_image_edit (free-form edits) or edit_image (upscale/reframe/removebg/enhance_skin/magic_edit). Do NOT call generate_image again.'
+            _followup_hint: 'If the user asks to edit/change/modify this image next (scene, lighting, objects, style, color — any content edit), pass urls[0] to generate_image_edit. Use edit_image ONLY for mechanical ops (upscale/reframe/removebg/enhance_skin). Do NOT call generate_image again.'
           }, null, 2)
         }, ...images]
       };
@@ -102,10 +102,10 @@ function registerGenerateTools(server, client, options = {}) {
   // ─── generate_image_edit ──────────────────────────────────
   server.tool(
     'generate_image_edit',
-    'Edit or transform an existing image using AI. Provide the source image URL(s) in `source_images` and describe the edit in `prompt` (e.g., "remove the background", "change the car color to red", "add sunglasses to the person"). Supports Visual DNA profiles and moodboards for style-consistent edits. For creating a brand new image from scratch, use generate_image. Returns the edited image URL(s) when complete.',
+    'THE tool for ANY prompt-driven / content edit of an existing image — changing the scene ("make it night", "change the sky to sunset"), adding/removing/replacing objects, restyling, recoloring, compositing, or any "edit this image to…" request. This is the image-editing equivalent of generate_image and runs on strong dedicated editing models (nano-banana-2, gpt-image-2). Provide the source image URL(s) in `source_images` and the instruction in `prompt`. Supports Visual DNA profiles and moodboards for style-consistent edits. Do NOT use `edit_image` for these — that tool is only for mechanical enhancements (upscale/reframe/remove-background/skin). For a brand-new image from scratch, use generate_image. Returns the edited image URL(s) when complete.',
     {
       prompt: z.string().describe('Description of the edit to apply (e.g., "remove the background", "change the sky to sunset")'),
-      model: z.string().optional().describe('Model identifier. Use list_models type="image_editing" to see options. Omit for Smart Select.'),
+      model: z.string().optional().describe('Model identifier — REQUIRED in practice: pick a specific editing model, do NOT omit (omitting = Smart Select auto-pick, which we avoid). Strong current default: "nano-banana-2" (best general prompt-driven editor) or "gpt-image-2" for photoreal edits. Call list_models type="image_editing" to see all options and pick per the user\'s intent.'),
       source_images: z.array(z.string()).describe('PIXEL-ACCURATE compositing. Array of source image URLs whose pixel content is composited into the output. **Cap: pass at most `max_reference_images` URLs from list_models for the chosen model — exceeding it is a deterministic 400.** Three modes the model auto-detects from input shape: (1) Single image → edit/transform that image. (2) Multiple images, one base + others → composite the others into the base. (3) Multiple images with no clear base → generate a new scene that pixel-accurately embeds the supplied images at positions described in the prompt. Mode 3 is the canonical pattern for thumbnails / branded compositions where exact-pixel logo + face fidelity matter. Refer to source images in the prompt by ordinal position ("FIRST source image", "SECOND source image") or use @image1/@image2 tags. Add "composite AS-IS, do not redraw or restyle" to lock pixels.'),
       aspect_ratio: z.string().optional().describe('Output aspect ratio (e.g., "1:1", "16:9", "9:16"). Must be in the chosen model\'s `supported_aspect_ratios` from list_models. Default: "1:1"'),
       enhance_prompt: z.boolean().optional().describe('Enhance the prompt for better results. Default: true'),
@@ -162,7 +162,7 @@ function registerGenerateTools(server, client, options = {}) {
     {
       prompt: z.string().describe('Creative brief or concept describing the full set of scenes to generate'),
       scene_count: z.number().optional().describe('Number of scenes/images to generate, 1–8. Default: 4. Use this — NOT num_images — to control how many outputs are created.'),
-      model: z.string().optional().describe('Model identifier applied to every scene. Omit for Smart Select.'),
+      model: z.string().optional().describe('Model identifier applied to every scene. Pick a SPECIFIC model — do NOT omit (omitting = Smart Select auto-pick, which we avoid); call list_models for this type and choose the model that best fits the user\'s intent.'),
       aspect_ratio: z.string().optional().describe('Aspect ratio applied to every scene (e.g., "1:1", "16:9", "9:16"). Must be in the chosen model\'s `supported_aspect_ratios` from list_models. Default: "1:1"'),
       workflow_type: z.string().optional().describe('"image" (default) or "video"'),
       duration: z.number().optional().describe('Duration in seconds per scene (video mode only). Must be a value in `supported_durations` from list_models, OR within `min_output_duration`-`max_output_duration`. E.g., 5 or 10.'),
@@ -598,7 +598,7 @@ function registerGenerateTools(server, client, options = {}) {
     'Generate a video from reference elements (images, videos, and/or audio) + a text prompt. Use when the user wants to animate specific uploaded/referenced assets — e.g. "animate this product", "put these 3 characters into a scene". IMPORTANT: different models accept different numbers of inputs — call list_models type="elements" and read elements_max_images / elements_max_videos / elements_max_audio on the chosen model before generating. For text-only → video use generate_video instead. For animating a single still image use generate_video_from_image. Returns the final video URL when complete.',
     {
       prompt: z.string().describe('Text description of the desired video / animation'),
-      model: z.string().optional().describe('Model identifier. Use list_models type="elements" to see options (Seedance 2, Kling O3 Reference, Grok Imagine, Veo 3.1, etc.). Check elements_max_images / elements_max_videos / elements_max_audio on the model. Omit for Smart Select.'),
+      model: z.string().optional().describe('Model identifier. Use list_models type="elements" to see options (Seedance 2, Kling O3 Reference, Grok Imagine, Veo 3.1, etc.). Check elements_max_images / elements_max_videos / elements_max_audio on the model. Pick a SPECIFIC model — do NOT omit (omitting = Smart Select auto-pick, which we avoid); call list_models for this type and choose the model that best fits the user\'s intent.'),
       reference_images: z.array(z.string()).optional().describe('Array of public image URLs used as reference elements (product shots, character references, etc.). **Cap: pass at most `elements_max_images` URLs from list_models for the chosen model — exceeding it is a deterministic 400.**'),
       reference_videos: z.array(z.string()).optional().describe('Array of reference video URLs for models that accept video inputs. **Cap: pass at most `elements_max_videos` URLs from list_models — if the cap is 0 the model rejects videos.**'),
       audio_url: z.string().optional().describe('URL of a reference audio track. **Audio constraints: `elements_max_audio` from list_models gates whether audio is accepted at all; audio duration must fall within `min_audio_duration`-`max_audio_duration`; format must be in `supported_audio_formats` (if specified).**'),
@@ -681,7 +681,7 @@ function registerGenerateTools(server, client, options = {}) {
       first_frame: z.string().optional().describe('URL or absolute local path to the first frame (file mode — alternative to first_frame_url)'),
       last_frame: z.string().optional().describe('URL or absolute local path to the last frame (file mode — alternative to last_frame_url)'),
       prompt: z.string().optional().describe('Optional description of the desired motion between the two frames (e.g. "smooth camera dolly in")'),
-      model: z.string().optional().describe('Model identifier. Use list_models type="firstlastgenerations" to see options. Omit for Smart Select.'),
+      model: z.string().optional().describe('Model identifier. Use list_models type="firstlastgenerations" to see options. Pick a SPECIFIC model — do NOT omit (omitting = Smart Select auto-pick, which we avoid); call list_models for this type and choose the model that best fits the user\'s intent.'),
       duration: z.number().optional().describe('Duration in seconds. Must be in `supported_durations` from list_models, OR within `min_output_duration`-`max_output_duration`. Default: 5'),
       aspect_ratio: z.string().optional().describe('Aspect ratio (auto-detected from first frame if not provided). Must be in `supported_aspect_ratios` from list_models when set. Default: "16:9"'),
       enhance_prompt: z.boolean().optional().describe('Enhance the prompt. Default: true'),
@@ -758,7 +758,7 @@ function registerGenerateTools(server, client, options = {}) {
       source: z.string().describe('URL or absolute local path to the source image or video (the face to animate). For lipsync-video: duration must fall within `min_video_duration`-`max_video_duration` from list_models.'),
       audio: z.string().describe('URL or absolute local path to the audio track (the voice to sync to). Duration must fall within `min_audio_duration`-`max_audio_duration` from list_models; format must be in `supported_audio_formats` (when set).'),
       text_prompt: z.string().optional().describe('Optional text prompt (for performance-capable models). For Sync-3 this is the free-text emotion/acting prompt, e.g. "speaking with excitement, calm and serious".'),
-      model: z.string().optional().describe('Model identifier. Use list_models type="lipsync-image" or type="lipsync-video" to see options. Omit for Smart Select.'),
+      model: z.string().optional().describe('Model identifier. Use list_models type="lipsync-image" or type="lipsync-video" to see options. Pick a SPECIFIC model — do NOT omit (omitting = Smart Select auto-pick, which we avoid); call list_models for this type and choose the model that best fits the user\'s intent.'),
       bounding_box_target: z.array(z.number()).optional().describe('Optional bounding box [x, y, w, h] for multi-face inputs (Hedra Character3 style). Leave empty for single-face.'),
       // Sync-3 (fal-ai/sync-lipsync/v3) only; ignored by other models.
       sync_mode: z.enum(['cut_off', 'loop', 'bounce', 'silence', 'remap']).optional().describe('Sync-3 / sync-lipsync family: how to reconcile an audio/video length mismatch. Default cut_off.'),
@@ -867,7 +867,7 @@ function registerGenerateTools(server, client, options = {}) {
     {
       source_video: z.string().describe('URL or absolute local path to the primary source video to restyle. **Source duration must fall within `min_video_duration`-`max_video_duration` from list_models for the chosen model** — videos outside that range are rejected (or silently truncated by some upstream providers). For models that use reference_videos as their primary input (e.g. WAN 2.6 reference-to-video), pass the first reference video here and also include it in reference_videos.'),
       prompt: z.string().optional().describe('Text description of the desired restyle / transformation. Required by most video-to-video models; omit for prompt-less models (VEED Subtitles, Act Two, Wan Animate, Kling Motion Control).'),
-      model: z.string().optional().describe('Model identifier. Use list_models type="video_to_video" to see options and check max_images / max_videos / max_elements / max_video_duration per model. Omit for Smart Select.'),
+      model: z.string().optional().describe('Model identifier. Use list_models type="video_to_video" to see options and check max_images / max_videos / max_elements / max_video_duration per model. Pick a SPECIFIC model — do NOT omit (omitting = Smart Select auto-pick, which we avoid); call list_models for this type and choose the model that best fits the user\'s intent.'),
       aspect_ratio: z.string().optional().describe('Output aspect ratio. Must be in `supported_aspect_ratios` from list_models when set. Default: matches source'),
       duration: z.number().optional().describe('Output duration in seconds. Must be in `supported_durations` from list_models, OR within `min_output_duration`-`max_output_duration`. Default: matches source'),
       enhance_prompt: z.boolean().optional().describe('Enhance the prompt. Default: true'),
@@ -1081,17 +1081,17 @@ function registerGenerateTools(server, client, options = {}) {
   // ─── edit_image ────────────────────────────────────────────
   server.tool(
     'edit_image',
-    'Apply a targeted AI edit to an existing image. Use for upscaling resolution, changing aspect ratio (reframe), removing the background, portrait skin enhancement, or a text-guided edit (magic_edit). Faster and cheaper than generate_image_edit for these specific operations because it routes to specialized models. Returns the edited image URL when complete.',
+    'Apply a MECHANICAL enhancement to an existing image: upscale resolution, reframe (change aspect ratio), remove background, or portrait skin retouching. ⚠️ For any PROMPT-DRIVEN / CONTENT edit — changing the scene, lighting, or time of day, adding/removing/replacing objects, restyling, recoloring — use `generate_image_edit` instead (stronger dedicated editing models, better results). Do NOT use this tool for "make it night / add sunglasses / change the background to X"-type edits. Returns the edited image URL when complete.',
     {
       image_url: z.string().describe('URL of the source image to edit'),
       operation: z.enum(['upscale', 'reframe', 'removebg', 'enhance_skin', 'magic_edit'])
-        .describe('Edit operation to apply: "upscale" (increase resolution 2×–4×), "reframe" (change aspect ratio), "removebg" (remove background), "enhance_skin" (portrait retouching), "magic_edit" (text-guided edit — requires prompt)'),
+        .describe('Mechanical edit operation: "upscale" (increase resolution 2×–4×), "reframe" (change aspect ratio), "removebg" (remove background), "enhance_skin" (portrait retouching). NOTE: "magic_edit" (text-guided content edit) is DEPRECATED here — use `generate_image_edit` for prompt-driven edits instead; it produces better results on stronger models.'),
       model: z.string().optional().describe('Model identifier override. Omit to use the default model for the operation.'),
       scale: z.number().optional().describe('Upscale factor: 2, 3, or 4. Only used when operation="upscale". Default: 2.'),
       aspect_ratio: z.string().optional().describe('Target aspect ratio (e.g., "16:9", "9:16", "1:1"). Required for operation="reframe".'),
       skin_strength: z.enum(['subtle', 'realistic', 'pimple', 'freckle']).optional()
         .describe('Skin enhancement style. Only used when operation="enhance_skin". Default: "realistic".'),
-      prompt: z.string().optional().describe('Text instruction for the edit. Required for operation="magic_edit" (e.g., "add sunglasses", "change the sky to sunset").'),
+      prompt: z.string().optional().describe('Text instruction — only for the deprecated operation="magic_edit". Prefer `generate_image_edit` for prompt-driven edits.'),
       project_id: projectIdField
     },
     async ({ image_url, operation, model, scale, aspect_ratio, skin_strength, prompt, project_id }) => {