@writepanda/mcp 1.214.0 → 1.216.0

package/bin/server.mjs

@@ -3154,7 +3154,7 @@ const TOOLS = [
 	{
 		name: "project_set_webcam_layout",
 		description:
-			"Project-level webcam layout, position, scale and crop. side-by-side: cx < 0.5 = camera left. Per-clip: project_set_clip_layout; vertical screen recordings: project_set_vertical_screen_layout.",
+			"Project-level webcam layout, position, scale and crop. cutout: presenter over the full screen with camera background removed, no box; scale=1 gives 55% frame height, cy=1 allows the lower half offscreen. side-by-side: cx < 0.5 = camera left. Per-clip: project_set_clip_layout; vertical screen recordings: project_set_vertical_screen_layout.",
 		inputSchema: {
 			type: "object",
 			properties: {
@@ -3163,6 +3163,7 @@ const TOOLS = [
 				preset: {
 					type: "string",
 					enum: [
+						"cutout",
 						"picture-in-picture",
 						"vertical-stack",
 						"side-by-side",
@@ -3756,6 +3757,7 @@ const TOOLS = [
 						"podcast",
 						"podcast-host-full",
 						"podcast-guest-full",
+						"cutout",
 						"picture-in-picture",
 						"side-by-side",
 						"vertical-stack",
@@ -3857,10 +3859,16 @@ const TOOLS = [
 	},
 	{
 		name: "transcript_delete_words",
-		description: "Delete words (their spans become trims).",
+		description:
+			"Delete words with measured quiet cut edges; cuts without pauses warn. snapToAudio=false uses legacy timing.",
 		inputSchema: {
 			type: "object",
 			properties: {
+				snapToAudio: {
+					type: "boolean",
+					description:
+						"Default true: measured quiet cuts; false uses legacy timing. Results include snapped/skipped/warnings.",
+				},
 				id: { type: "string" },
 				path: { type: "string" },
 				wordIds: { type: "array", items: { type: "string" } },
@@ -3905,10 +3913,16 @@ const TOOLS = [
 	},
 	{
 		name: "transcript_remove_fillers",
-		description: "Trim filler words (um, uh, hmm) and immediate repeats.",
+		description:
+			"Trim fillers and repeats only when measured pauses surround them; skipped fillers intentionally remain. Returns snapped/skipped/warnings; snapToAudio=false opts out.",
 		inputSchema: {
 			type: "object",
 			properties: {
+				snapToAudio: {
+					type: "boolean",
+					description:
+						"Default true: measured quiet cuts; false uses legacy timing. Results include snapped/skipped/warnings.",
+				},
 				id: { type: "string" },
 				path: { type: "string" },
 				aggressive: {
@@ -3974,14 +3988,22 @@ const TOOLS = [
 	{
 		name: "transcript_remove_silences",
 		description:
-			"Trim silences (transcript gaps plus audio dead air). Returns { removedCount, totalTrimmedMs }.",
+			"Remove measured quiet runs with a 50ms decay hold. Returns snapped/skipped/warnings; snapToAudio=false uses legacy timing.",
 		inputSchema: {
 			type: "object",
 			properties: {
+				snapToAudio: {
+					type: "boolean",
+					description:
+						"Default true: measured quiet cuts; false uses legacy timing. Results include snapped/skipped/warnings.",
+				},
 				id: { type: "string" },
 				path: { type: "string" },
 				thresholdMs: { type: "number", description: "Default 600." },
-				paddingMs: { type: "number", description: "Kept around words, default 100." },
+				paddingMs: {
+					type: "number",
+					description: "Legacy/fallback word guard, default 100. Measured audio uses a 50ms hold.",
+				},
 				expectedRevision: { type: "number" },
 			},
 		},
@@ -5783,7 +5805,7 @@ Before an editing pass, read the skill's "Which tool for which moment" section (
 4. transcript_find_issues, then transcript_delete_words on each duplicate take / false start (keep the LAST take; keep severity "low" ones).
 5. transcript_remove_silences.
 6. audio_clean clips not yet cleaned.
-7. The style, after steps 2-5 (they shift the timeline). The user's named style, recipe, creator or a saved memory wins. Otherwise call recipe_pick: it returns the recipes that fit this project's format and footage, each with pickWhen, plus defaultId. Read the transcript, choose the ONE candidate whose pickWhen matches what is said (defaultId when none clearly fits), do its "before" step if any, then recipe_apply_style + recipe_render and follow the prompt: the recipe decides captions, graphics, zooms, music and transitions. (An app without recipe_pick, 2.0.5 and older: recipe_list with format short or long, and choose by description and footage.) Tell the user which recipe and why in one line and name one or two alternatives. Only when no candidate exists or the user asked for a plain clean-up: captions (caption_toggle, template glowStack), a few motion graphics, emphasis zooms.
+7. The style, after steps 2-5 (they shift the timeline). The user's named style, recipe, creator or a saved memory wins. Otherwise call recipe_pick: it returns the recipes that fit this project's format and footage, each with pickWhen, plus defaultId. Read the transcript, choose the ONE candidate whose pickWhen matches what is said (defaultId when none clearly fits), do its "before" step if any, then recipe_apply_style + recipe_render and follow the prompt: the recipe decides the style (captions, pacing, kinds of moves, music, transitions). Graphics inside it are designed by you: a named template is the reference look; author your own (motion_render_html, same palette and type) when a beat needs something the template can't show. (An app without recipe_pick, 2.0.5 and older: recipe_list with format short or long, and choose by description and footage.) Tell the user which recipe and why in one line and name one or two alternatives. Only when no candidate exists or the user asked for a plain clean-up: captions (caption_toggle, template glowStack), a few motion graphics, emphasis zooms.
 8. B-roll, last: pick the moments yourself from what's said (any language), check them with project_check_broll, show the user your moments and ASK: their own images, generated on their image connector (only if one is connected; it uses their credits), or no B-roll. Never generate or invent images without their answer. Place what they give with project_add_broll. Nobody to ask: skip it and say so.
 9. llm_generate_title / description / timestamps, then preview_show and ask before export_start. Never export before the user has seen the preview.
 Report the real count each step returned. No music, intro/outro cards or FX overlays unless the recipe or the user asks for them.

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@writepanda/mcp",
-	"version": "1.214.0",
+	"version": "1.216.0",
 	"description": "Model Context Protocol server for PandaStudio. Exposes the desktop video editor's automation surface to Cursor, Continue, Cline, Claude Desktop, and any MCP-compliant client.",
 	"keywords": [
 		"pandastudio",