@kolbo/kolbo-code-linux-arm64-musl 1.1.71 → 1.1.73

package/skills/typography-video/reference/KineticTitleScene.tsx

@@ -0,0 +1,345 @@
+/**
+ * KineticTitleScene — Full-screen kinetic text card
+ *
+ * bgStyle="solid"         — always-on dark background (default)
+ * bgStyle="dynamic_panel" — bg wipes in before text, wipes out after text gone
+ * bgStyle="transparent"   — no background (composite over footage, render with alpha codec)
+ *
+ * ALL words land simultaneously (staggered by 4 frames each).
+ * Alternating solid / outline / accent treatment.
+ * Based on Higgsfield / SOUL 2.0 reference style.
+ */
+import React from "react";
+import { useCurrentFrame, useVideoConfig, spring, interpolate, Easing } from "remotion";
+import { loadFont as loadPoppins } from "@remotion/google-fonts/Poppins";
+import { loadFont as loadHeebo } from "@remotion/google-fonts/Heebo";
+
+const { fontFamily: poppins } = loadPoppins();
+const { fontFamily: heebo } = loadHeebo();
+
+export interface KineticTitleProps {
+  words: string[];           // 2–4 words, each on its own line
+  subtext?: string;          // optional small line below
+  accentColor?: string;      // default brand blue
+  language?: string;         // "en" | "he"
+  bgStyle?: "solid" | "dynamic_panel" | "transparent";
+  durationInFrames: number;
+  fps: number;
+}
+
+const BRAND_BLUE = "#60a5fa";
+const STAGGER = 4; // frames between each word entrance
+
+// How many frames the bg leads the text on entry, and trails it on exit
+const BG_LEAD = 10;   // bg starts wiping in 10 frames before text
+const BG_TRAIL = 12;  // bg stays for 12 frames after text is gone
+
+export const KineticTitleScene: React.FC<KineticTitleProps> = ({
+  words = ["WENT", "VIRAL"],
+  subtext,
+  accentColor = BRAND_BLUE,
+  language = "en",
+  bgStyle = "solid",
+  durationInFrames = 150,
+  fps = 30,
+}) => {
+  const frame = useCurrentFrame();
+  const { width, height } = useVideoConfig();
+
+  const isHebrew = language === "he" || language === "iw";
+  const fontFamily = isHebrew ? heebo : poppins;
+  const isVertical = height > width;
+  const accent = accentColor;
+  const isDynamic = bgStyle === "dynamic_panel";
+  const isTransparent = bgStyle === "transparent";
+
+  // ── Text timing ──────────────────────────────────────────────────────────
+  // Dynamic: text starts after bg has partially appeared
+  const textOffset = isDynamic ? BG_LEAD : 0;
+  const textExitStart = durationInFrames - (isDynamic ? BG_TRAIL + 12 : 12);
+
+  const exitOpacity = interpolate(
+    frame,
+    [textExitStart, textExitStart + 12],
+    [1, 0],
+    { extrapolateRight: "clamp" }
+  );
+
+  // Global entrance scale
+  const globalScale = interpolate(frame - textOffset, [0, 20], [1.04, 1.0], {
+    extrapolateLeft: "clamp",
+    extrapolateRight: "clamp",
+    easing: Easing.out(Easing.cubic),
+  });
+
+  // ── Background panel timing ───────────────────────────────────────────────
+  // BG enters: wipes left→right starting at frame 0
+  const bgEnterDuration = 18;
+  const bgEnterProgress = interpolate(frame, [0, bgEnterDuration], [0, 1], {
+    extrapolateLeft: "clamp",
+    extrapolateRight: "clamp",
+    easing: Easing.out(Easing.cubic),
+  });
+
+  // BG exits: wipes left→right starting after text is fully gone
+  const bgExitStart = textExitStart + 12;
+  const bgExitProgress = interpolate(
+    frame,
+    [bgExitStart, bgExitStart + 14],
+    [0, 1],
+    {
+      extrapolateLeft: "clamp",
+      extrapolateRight: "clamp",
+      easing: Easing.in(Easing.cubic),
+    }
+  );
+
+  // scaleX wipe — most reliable in Remotion headless Chrome
+  // Enter: scaleX grows 0→1 from LEFT origin  (reveals left to right)
+  // Exit:  scaleX shrinks 1→0 from RIGHT origin (removes left to right)
+  const bgScaleX = isDynamic
+    ? (frame < bgExitStart
+        ? interpolate(bgEnterProgress, [0, 1], [0, 1])
+        : interpolate(bgExitProgress,  [0, 1], [1, 0]))
+    : 1;
+  const bgTransformOrigin = isDynamic && frame >= bgExitStart ? "right center" : "left center";
+
+  const showBg = !isTransparent;
+
+  // ── Accent scan line across bg (cinematic HUD feel) ──────────────────────
+  const scanY = interpolate(frame, [0, durationInFrames], [0, height], {
+    extrapolateRight: "clamp",
+  });
+  const scanOpacity = isDynamic
+    ? bgEnterProgress * (1 - bgExitProgress) * 0.12
+    : 0.06;
+
+  // ── Font size ─────────────────────────────────────────────────────────────
+  const baseFontSize = isVertical
+    ? Math.round(width * 0.22)
+    : Math.round(height * 0.26);
+
+  // ── Word styles: solid → outline → accent ────────────────────────────────
+  const wordStyles = (i: number) => {
+    const mod = i % 3;
+    if (mod === 0) return { color: "#ffffff",   stroke: "none",  shadow: `0 0 40px ${accent}44` };
+    if (mod === 1) return { color: "transparent", stroke: accent, shadow: `0 0 30px ${accent}44` };
+    return                 { color: accent,     stroke: "none",  shadow: `0 0 50px ${accent}88` };
+  };
+
+  // ── Separator line ────────────────────────────────────────────────────────
+  const SepLine = ({ delay }: { delay: number }) => {
+    const lineW = interpolate(frame - delay, [0, 16], [0, 1], {
+      extrapolateRight: "clamp",
+      easing: Easing.out(Easing.cubic),
+    });
+    return (
+      <div style={{
+        width: `${lineW * 100}%`,
+        height: 1,
+        background: `linear-gradient(${isHebrew ? "270deg" : "90deg"}, transparent, ${accent}88, transparent)`,
+        marginBottom: 2,
+        marginTop: 2,
+      }} />
+    );
+  };
+
+  return (
+    <div style={{
+      width, height,
+      // dynamic_panel: outer is transparent — the animated inner panel owns the background
+      background: (isTransparent || isDynamic) ? "transparent" : "#07070f",
+      overflow: "hidden",
+      position: "relative",
+      fontFamily,
+      direction: isHebrew ? "rtl" : "ltr",
+    }}>
+
+      {/* ── Background panel (with clip-path wipe) ── */}
+      {showBg && (
+        <div style={{
+          position: "absolute", inset: 0,
+          transform: isDynamic ? `scaleX(${bgScaleX})` : undefined,
+          transformOrigin: isDynamic ? bgTransformOrigin : undefined,
+        }}>
+          {/* Dark fill */}
+          <div style={{ position: "absolute", inset: 0, background: "#07070f" }} />
+
+          {/* Grid lines */}
+          <div style={{
+            position: "absolute", inset: 0,
+            backgroundImage: [
+              `linear-gradient(rgba(255,255,255,0.03) 1px, transparent 1px)`,
+              `linear-gradient(90deg, rgba(255,255,255,0.03) 1px, transparent 1px)`,
+            ].join(", "),
+            backgroundSize: "60px 60px",
+            pointerEvents: "none",
+          }} />
+
+          {/* Diagonal corner accent */}
+          <div style={{
+            position: "absolute",
+            top: -60,
+            left: isHebrew ? undefined : -60,
+            right: isHebrew ? -60 : undefined,
+            width: 300, height: 300,
+            background: `linear-gradient(${isHebrew ? "225deg" : "135deg"}, ${accent}18 0%, transparent 60%)`,
+            pointerEvents: "none",
+          }} />
+
+          {/* Vignette */}
+          <div style={{
+            position: "absolute", inset: 0,
+            background: "radial-gradient(ellipse at center, transparent 30%, rgba(0,0,0,0.65) 100%)",
+            pointerEvents: "none",
+          }} />
+
+          {/* Horizontal scan line (slow drift) */}
+          <div style={{
+            position: "absolute",
+            left: 0, right: 0,
+            top: scanY,
+            height: 1,
+            background: `linear-gradient(90deg, transparent, ${accent}, transparent)`,
+            opacity: scanOpacity,
+            pointerEvents: "none",
+          }} />
+
+          {/* Bright edge line at wipe front — sits at the right edge of the scaled panel */}
+          {isDynamic && bgScaleX > 0 && bgScaleX < 1 && (
+            <div style={{
+              position: "absolute",
+              top: 0, bottom: 0,
+              right: 0,
+              width: Math.round(2 / Math.max(bgScaleX, 0.05)), // compensate for scaleX compression
+              background: `linear-gradient(180deg, transparent, ${accent}, ${accent}, transparent)`,
+              opacity: 0.9,
+              boxShadow: `0 0 16px ${accent}, 0 0 32px ${accent}88`,
+            }} />
+          )}
+        </div>
+      )}
+
+      {/* ── Word stack ── */}
+      <div style={{
+        position: "absolute",
+        inset: 0,
+        display: "flex",
+        flexDirection: "column",
+        justifyContent: "center",
+        paddingLeft:  isHebrew ? 0 : isVertical ? 32 : 64,
+        paddingRight: isHebrew ? (isVertical ? 32 : 64) : 0,
+        paddingTop: isVertical ? 40 : 30,
+        transform: `scale(${globalScale})`,
+        opacity: exitOpacity,
+      }}>
+        {words.map((word, i) => {
+          const delay = textOffset + i * STAGGER;
+          const wordScale = spring({
+            frame: Math.max(0, frame - delay),
+            fps,
+            from: 0,
+            to: 1,
+            config: { damping: 9, stiffness: 350 },
+          });
+          const wordY = interpolate(
+            frame - delay, [0, 14], [40, 0],
+            { extrapolateLeft: "clamp", extrapolateRight: "clamp", easing: Easing.out(Easing.cubic) }
+          );
+          const ws = wordStyles(i);
+
+          const glitchAmt = i === 0
+            ? interpolate(frame - textOffset, [0, 1, 3], [12, 4, 0], { extrapolateRight: "clamp" })
+            : 0;
+
+          return (
+            <React.Fragment key={i}>
+              {i > 0 && <SepLine delay={textOffset + (i - 1) * STAGGER + 6} />}
+              <div style={{
+                position: "relative",
+                transform: `translateY(${wordY}px) scaleY(${wordScale})`,
+                transformOrigin: isHebrew ? "top right" : "top left",
+                lineHeight: 0.85,
+                overflow: "visible",
+              }}>
+                {glitchAmt > 0 && (
+                  <div style={{
+                    position: "absolute",
+                    fontSize: baseFontSize,
+                    fontWeight: 900,
+                    color: accent,
+                    textTransform: isHebrew ? "none" : "uppercase",
+                    letterSpacing: isHebrew ? 0 : -2,
+                    transform: `translateX(${glitchAmt}px)`,
+                    opacity: 0.45,
+                    mixBlendMode: "screen",
+                    userSelect: "none",
+                  }}>{word}</div>
+                )}
+                <div style={{
+                  fontSize: baseFontSize,
+                  fontWeight: 900,
+                  color: ws.color,
+                  textTransform: isHebrew ? "none" : "uppercase",
+                  letterSpacing: isHebrew ? 0 : -2,
+                  WebkitTextStroke: ws.stroke !== "none" ? `4px ${ws.stroke}` : undefined,
+                  textShadow: ws.shadow !== "none" ? ws.shadow : undefined,
+                  whiteSpace: "nowrap",
+                  userSelect: "none",
+                }}>{word}</div>
+              </div>
+            </React.Fragment>
+          );
+        })}
+
+        {subtext && (() => {
+          const stDelay = textOffset + words.length * STAGGER + 10;
+          const stOpacity = interpolate(frame - stDelay, [0, 14], [0, 1], {
+            extrapolateLeft: "clamp", extrapolateRight: "clamp",
+          });
+          return (
+            <div style={{
+              fontSize: Math.round(baseFontSize * 0.17),
+              fontWeight: 600,
+              color: "rgba(255,255,255,0.45)",
+              textTransform: isHebrew ? "none" : "uppercase",
+              letterSpacing: isHebrew ? 0 : 5,
+              marginTop: 20,
+              opacity: stOpacity * exitOpacity,
+            }}>{subtext}</div>
+          );
+        })()}
+      </div>
+
+      {/* ── Viewfinder corners ── */}
+      {Math.floor(frame / 10) % 2 === 0 && [
+        { top: 24, left: 24,  borderTopWidth: 2, borderLeftWidth: 2  },
+        { top: 24, right: 24, borderTopWidth: 2, borderRightWidth: 2 },
+        { bottom: 24, left: 24,  borderBottomWidth: 2, borderLeftWidth: 2  },
+        { bottom: 24, right: 24, borderBottomWidth: 2, borderRightWidth: 2 },
+      ].map((pos, i) => (
+        <div key={i} style={{
+          position: "absolute", width: 20, height: 20,
+          borderColor: accent, borderStyle: "solid", borderWidth: 0,
+          opacity: isTransparent ? exitOpacity * 0.5 : 0.5,
+          ...pos,
+        }} />
+      ))}
+
+      {/* ── Frame counter ── */}
+      <div style={{
+        position: "absolute",
+        bottom: isVertical ? 32 : 24,
+        right: isHebrew ? undefined : 36,
+        left: isHebrew ? 36 : undefined,
+        fontSize: 11,
+        fontFamily: "monospace",
+        color: `${accent}66`,
+        letterSpacing: 3,
+        opacity: exitOpacity,
+      }}>
+        {String(frame).padStart(4, "0")}
+      </div>
+    </div>
+  );
+};

package/skills/video-editing/SKILL.md

@@ -0,0 +1,128 @@
+---
+name: video-editing
+description: >
+  Video editing decisions: what to cut vs keep, cut techniques (J-cut, L-cut, hard cut),
+  pacing by format, edit decision structure, silence/filler removal, talking head editing.
+  Use when making editorial decisions about video content.
+  Keywords: video editing, cut, trim, j-cut, l-cut, pacing, filler words, silence, talking head,
+  edit decision, transition, dead air, false start
+---
+
+# Video Editing — Editorial Decisions
+
+## What to Cut
+
+1. **Filler words:** "um", "uh", "like", "you know" — cut at word boundaries using word timestamps
+2. **False starts:** When the speaker restarts a sentence, keep only the final take
+3. **Dead air:** Silence longer than 1.5 seconds should be trimmed to ~0.5 seconds
+4. **Off-topic tangents:** If the speaker wanders, cut to the next relevant segment
+5. **Repeated points:** Keep the best delivery, remove redundant takes
+
+## What NOT to Cut
+
+- **Breath pauses:** Natural 0.3-0.8 second pauses between sentences. These sound natural.
+- **Emphasis pauses:** Intentional pauses for dramatic effect
+- **Reactions and transitions:** Verbal bridges like "So..." or "Now..." that provide flow
+
+## Cut Techniques
+
+| Technique | Description | When to Use |
+|-----------|-------------|-------------|
+| **J-cut** | Audio from next segment starts ~0.5s before visual cut | Smooth transitions |
+| **L-cut** | Audio from current segment continues ~0.5s after visual cut | Maintaining continuity |
+| **Hard cut** | Instant transition | Major topic changes |
+| **Jump cut** | Cut within same shot (visible jump) | YouTube/social energy, pacing |
+| **Match cut** | Visual similarity bridges two different shots | Creative storytelling |
+
+## Pacing by Format
+
+| Format | Approach |
+|--------|----------|
+| **Short-form (< 60s)** | Aggressive cuts. Minimal dead air. High energy. Visual change every 1-3s |
+| **Medium-form (1-10 min)** | Balanced. Keep natural pauses for breathing room. Change every 3-5s |
+| **Long-form (> 10 min)** | Let scenes breathe. Only cut obvious problems. Change every 5-10s |
+
+## Edit Decision Structure
+
+When planning an edit, define:
+
+- **Cuts:** Ordered list of segments to keep (source, in/out points, speed)
+- **Overlays:** Timed overlay placements (images, diagrams, lower thirds)
+- **Subtitles:** Subtitle configuration (enabled, style, source file)
+- **Music:** Background music settings (asset, volume, ducking, fades)
+- **Transitions:** Transition type and timing between cuts
+
+## Silence Removal Workflow
+
+1. **Detect silence** with FFmpeg: `silencedetect=noise=-35dB:d=0.4`
+2. **Parse** silence_start/silence_end timestamps from stderr
+3. **Generate segments** of non-silent audio
+4. **Concatenate** segments with the concat demuxer
+5. Optional: apply `atempo=1.14` for subtle speedup that feels natural
+
+## Talking Head Editing Checklist
+
+- [ ] No visible jump cuts without intentional style choice
+- [ ] Audio doesn't pop or click at cut points
+- [ ] Pacing matches content energy and target platform
+- [ ] Speaker's face is never covered by overlays
+- [ ] All cuts are at word boundaries (not mid-word)
+- [ ] Filler words removed unless they serve the personality
+- [ ] B-roll covers any remaining jump cuts
+
+## Lip Sync (Dubbing / Localization)
+
+When replacing audio and matching lips:
+
+| Input | Tool | Output |
+|-------|------|--------|
+| Existing VIDEO + new audio | Lip sync | Video with synced lips |
+| Still PHOTO + audio | Talking head generator | New video from photo |
+
+**Decision rule:** If you have video footage of the person, use lip sync. If you only have a photo, use talking head generation.
+
+**Workflow for localization:**
+```
+transcribe(video) → translate → TTS(translated text) → lip_sync(original_video, new_audio)
+```
+
+Keep original video as source for each language — never chain lip sync outputs.
+
+**Face padding** for lip sync: `[0, 10, 0, 0]` (top, bottom, left, right) works for 90% of footage. Increase bottom if chin gets cropped.
+
+---
+
+## Kolbo MCP Integration
+
+| Task | Kolbo MCP Tool | Notes |
+|------|---------------|-------|
+| Transcribe for edit points | `transcribe_audio` | Word-level timestamps for precise cuts |
+| Lip sync dubbing | `generate_lipsync` | Source video + new audio |
+| Generate B-roll | `generate_video` or `generate_image` | Cover jump cuts |
+| Generate narration | `generate_speech` | Re-record with AI voice |
+| Visual analysis | `chat_send_message` + Gemini | "Analyze this video for edit points" |
+
+**Editing workflow with Kolbo:**
+1. `transcribe_audio` → get full transcript with word timestamps
+2. Identify filler words, dead air, false starts from transcript
+3. Generate FFmpeg trim commands for non-silent/non-filler segments
+4. `generate_image` or `generate_video` → B-roll for covering jump cuts
+5. Concatenate clips + burn-in subtitles + mix audio
+6. Review with `production-review` skill
+
+**Localization workflow:**
+1. `transcribe_audio` → source language transcript
+2. Translate the transcript (use `chat_send_message` for translation)
+3. `generate_speech` → TTS in target language
+4. `generate_lipsync` → sync new audio to original face
+5. Repeat for each language (always from original, never chain)
+
+---
+
+## Local / Free Options
+
+> **IMPORTANT:** Always use Kolbo MCP tools by default (`transcribe_audio`, `generate_lipsync`). FFmpeg silence removal is safe to use directly. For anything else, confirm with the user first.
+
+**FFmpeg (safe, standard):** Silence detection/removal, trimming, concatenation — all built-in.
+
+**Transcription:** If the user wants offline, `faster-whisper` runs on CPU (`pip install faster-whisper`). Confirm before installing.

package/skills/video-production/SKILL.md

@@ -105,16 +105,15 @@ def detect_silence(video_path, noise_db=-35, duration=0.4):
 
 ## RTL (Hebrew/Arabic) Subtitles
 
-SRT with `subtitles` filter works for basic burn. For per-word karaoke highlighting with RTL:
+For comprehensive RTL subtitle handling, load the `subtitle-production` skill — it contains full patterns for:
+- Simple SRT burn-in with Heebo font + `Encoding=177`
+- ASS per-word positioning for karaoke (with PIL `~0.74` scale factor)
+- Remotion RTL captions with CSS `direction: rtl` and all the flip rules
+- RTL progress bar with FFmpeg `geq` filter
 
-- Each word gets its own ASS `Dialogue` line with explicit `\pos(x,y)`
-- Use PIL to measure word widths: apply `~0.74` scale factor (PIL→libass calibration)
-- Use `Alignment=7` (top-left anchor) so `\pos` sets exact top-left of each word
-- Set `Encoding=177` (Hebrew) in ASS style
-- Strip punctuation and render as separate positioned elements
-- Two ASS styles (e.g., White + Yellow) instead of inline `\c` color tags
+**CRITICAL**: Any inline ASS tag (`\c`, `\K`, `\1c`, etc.) between RTL words breaks Unicode bidi in libass — words render LTR. Use separate Dialogue lines per word instead.
 
-**CRITICAL**: Any inline ASS tag (`\c`, `\K`, `\1c`, etc.) between RTL words breaks Unicode bidirectional algorithm in libass — words render LTR. Use separate Dialogue lines instead.
+For Remotion RTL layout rules (padding flips, transform-origin, gradient direction), load the `typography-video` skill.
 
 ## Remotion Motion Graphics