davinci-resolve-mcp 2.187.0 → 2.189.0

package/CHANGELOG.md

@@ -2,6 +2,48 @@
 
 Release history for the DaVinci Resolve MCP Server. The latest release is summarized in the root README; older entries live here to keep the README focused.
 
+## What's New in v2.189.0 — E136: Premiere markers belong to their sequence and read their real fields
+
+### Fixed
+
+- **Every project marker landed on every sequence, at frame 0, unnamed.**
+  The marker walk returned every `Marker` object in the project for each
+  sequence and read fields the real shape does not carry: on a real Premiere
+  2025 turnover that was 1,228 bogus markers per sequence. A sequence's
+  markers are its own — `Sequence.MarkerOwner.Markers` → a markers container
+  → `<Marker>` pairs → marker objects whose payload is a `DVAMarker` JSON
+  blob (`mStartTime.ticks`, `mName`, `mComment`, `mType`, `mEndTime`). They
+  now read from that blob (ticks to frames, duration from an end time),
+  sorted by frame; the legacy field shape still reads when a sequence owns
+  it; a sequence without a marker owner reports none. On the turnover: 188
+  markers across 66 sequences, where every sequence used to show 1,228.
+
+### Measured (filed in api-limitations)
+
+- The Premiere 2025 marker ownership chain and the `DVAMarker` JSON payload.
+
+## What's New in v2.188.0 — E135: Premiere tracks are lanes; nested blocks first-fit; an audio lane ceiling
+
+### Fixed
+
+- **Every Premiere track was labelled `V` or `A`.** Multi-track sequences
+  collapsed onto one lane: on a real Premiere 2025 reels project 687 of 741
+  sequences had cuts overlapping on the same lane (4,631 pairs), which the
+  bridge refuses. Tracks now number per kind in track order (`V`, `V2` … /
+  `A`, `A2` …) across track groups and the legacy lists, and clip events,
+  fade legs and transitions carry the lane.
+- **Nested sequences expand in a second pass, first-fit.** Two nested
+  sequences stacked on different parent lanes each bring their own inner
+  lanes, and a fixed lane offset collided on 7,615 pairs. A nested block now
+  expands after every parent lane is known and shifts up by the smallest
+  offset at which none of its cuts overlap what the parent or an earlier
+  block holds (`laneShift` records it). On the reels project the only
+  "overlaps" left are zero-length fade carriers — none real.
+- **An audio lane ceiling in the bridge.** Flattened nested sequences can
+  stack audio lanes to A40 (3,586 events above lane 16 on that project, in
+  30 sequences); Resolve authors A1–A16. Events above 16 now drop with a
+  reason in `audioLanesBeyondCeiling` instead of authoring blind.
+
 ## What's New in v2.187.0 — E134: real Premiere transitions attach, with their real type
 
 ### Fixed

package/README.md

@@ -2,7 +2,7 @@
 
 English | [简体中文](README.zh-CN.md)
 
-[![Version](https://img.shields.io/badge/version-2.187.0-blue.svg)](https://github.com/samuelgursky/davinci-resolve-mcp/releases)
+[![Version](https://img.shields.io/badge/version-2.189.0-blue.svg)](https://github.com/samuelgursky/davinci-resolve-mcp/releases)
 [![npm](https://img.shields.io/npm/v/davinci-resolve-mcp.svg?label=npm&color=CB3837)](https://www.npmjs.com/package/davinci-resolve-mcp)
 [![API Coverage](https://img.shields.io/badge/API%20Coverage-100%25-brightgreen.svg)](docs/reference/api-coverage.md)
 [![Tools](https://img.shields.io/badge/MCP%20Tools-36%20(353%20full)-blue.svg)](#server-modes)

package/README.zh-CN.md

@@ -2,7 +2,7 @@
 
 [English](README.md) | 简体中文
 
-[![Version](https://img.shields.io/badge/version-2.187.0-blue.svg)](https://github.com/samuelgursky/davinci-resolve-mcp/releases)
+[![Version](https://img.shields.io/badge/version-2.189.0-blue.svg)](https://github.com/samuelgursky/davinci-resolve-mcp/releases)
 [![npm](https://img.shields.io/npm/v/davinci-resolve-mcp.svg?label=npm&color=CB3837)](https://www.npmjs.com/package/davinci-resolve-mcp)
 [![API Coverage](https://img.shields.io/badge/API%20Coverage-100%25-brightgreen.svg)](docs/reference/api-coverage.md)
 [![Tools](https://img.shields.io/badge/MCP%20Tools-36%20(353%20full)-blue.svg)](#服务器模式)
@@ -12,7 +12,7 @@
 [![Python](https://img.shields.io/badge/python-3.10+-green.svg)](https://www.python.org/downloads/)
 [![License](https://img.shields.io/badge/license-MIT-blue.svg)](https://opensource.org/licenses/MIT)
 
-> 本翻译对应 v2.187.0 版 README。如与英文原版有出入，以 [英文原版](README.md) 为准。
+> 本翻译对应 v2.189.0 版 README。如与英文原版有出入，以 [英文原版](README.md) 为准。
 
 一个 Model Context Protocol (MCP) 服务器，让 AI 助手通过官方脚本 API 控制 DaVinci Resolve Studio（达芬奇）。它提供完整的 API 覆盖，外加带护栏的工作流助手，涵盖剪辑、媒体池整理、渲染设置、审阅标记、调色、Fusion、Fairlight、项目生命周期任务、扩展开发，以及不碰源媒体的媒体分析。
 

package/docs/reference/api-limitations.md

@@ -322,7 +322,7 @@ equivalent, blocking full automation.
 - **Object:** `editorial.parse_interchange / list_sequences (prproj)`
 - **Signature:** `(path.prproj) -> sequences/events`
 - **Behavior:** Measured on a real 130 MB colour turnover (E132): objects are defined in TWO id spaces — numeric ObjectID referenced by ObjectRef, and uuid ObjectUID referenced by ObjectURef — and Sequence, ClipProjectItem, MasterClip, Media, VideoClipTrack and AudioClipTrack are all UID-defined. A Sequence lists its tracks through <TrackGroups><TrackGroup><Second ObjectRef> → VideoTrackGroup / AudioTrackGroup (a third DataTrackGroup is empty) → <TrackGroup><Tracks><Track ObjectURef> → Video/AudioClipTrack → <ClipTrack><ClipItems><TrackItems><TrackItem ObjectRef>. An item's record span sits under <ClipTrackItem><TrackItem> (Start/End ticks) and its source under ClipTrackItem.SubClip → SubClip.Clip → VideoClip.Clip (InPoint/OutPoint, Source) → VideoMediaSource.MediaSource.Media (ObjectURef) → Media (FilePath / ActualMediaFilePath); SubClip.Name and MasterClip.Name are the fallbacks. A ZERO is written as ABSENCE (the counting leader at record 0 has End and no Start). Names are direct <Name> children. Legacy/synthetic files instead use ObjectID/ObjectRef throughout, <VideoTracks>/<AudioTracks> track lists, Start/End/InPoint/OutPoint on the item and Node/Properties names.
-- **Workaround / current handling:** Index both id spaces and follow both reference attributes; walk TrackGroups when present; read timing through the *TrackItem child; treat a missing Start/InPoint beside a present End/OutPoint as 0. With that, the turnover lists all 739 sequences and its reel walks 335 events (it listed ZERO before). A NESTED SEQUENCE used as a clip has Clip.Source → VideoSequenceSource / AudioSequenceSource → SequenceSource.Sequence ObjectURef → the nested Sequence (E133: 3607 such items in the reels project); parse_interchange flattens them through the clip's InPoint window with fromCompound. TRANSITIONS sit in a SEPARATE track list, ClipTrack.TransitionItems.TrackItems (E134): a Video/AudioTransitionTrackItem carries TransitionTrackItem.TrackItem{Start,End} (the span), DisplayName / MatchName (the effect), Alignment (ticks) and HasIncomingClip / HasOutgoingClip — false marks a fade from/to black or silence.
+- **Workaround / current handling:** Index both id spaces and follow both reference attributes; walk TrackGroups when present; read timing through the *TrackItem child; treat a missing Start/InPoint beside a present End/OutPoint as 0. With that, the turnover lists all 739 sequences and its reel walks 335 events (it listed ZERO before). A NESTED SEQUENCE used as a clip has Clip.Source → VideoSequenceSource / AudioSequenceSource → SequenceSource.Sequence ObjectURef → the nested Sequence (E133: 3607 such items in the reels project); parse_interchange flattens them through the clip's InPoint window with fromCompound. TRANSITIONS sit in a SEPARATE track list, ClipTrack.TransitionItems.TrackItems (E134): a Video/AudioTransitionTrackItem carries TransitionTrackItem.TrackItem{Start,End} (the span), DisplayName / MatchName (the effect), Alignment (ticks) and HasIncomingClip / HasOutgoingClip — false marks a fade from/to black or silence. MARKERS (E136): a sequence owns its markers through Sequence.MarkerOwner.Markers → a Markers container → <Markers><Marker><Second ObjectRef> → a Marker object whose payload is a DVAMarker JSON string ({mMarkerID, mStartTime.ticks, mType, mName?, mComment?, mEndTime?}); clips own theirs the same way via Clip.MarkerOwner. Marker objects exist project-wide (1228 on the turnover) — only the owner chain says whose they are.
 - **Tags:** prproj, premiere, interchange, silent-failure
 
 ### ProjectManager.CreateProject (discards an unsaved current project)

package/install.py

@@ -37,7 +37,7 @@ from src.utils.update_check import (
 
 # ─── Version ──────────────────────────────────────────────────────────────────
 
-VERSION = "2.187.0"
+VERSION = "2.189.0"
 # Only hard floor: mcp[cli] requires Python 3.10+. There is no upper bound —
 # Resolve's scripting bridge loads into newer interpreters on recent builds
 # (Python 3.14 verified against Resolve Studio 20.3.2). Older Resolve builds

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "davinci-resolve-mcp",
-  "version": "2.187.0",
+  "version": "2.189.0",
   "description": "NPM bootstrapper for the DaVinci Resolve MCP Server.",
   "license": "MIT",
   "author": "Samuel Gursky <samgursky@gmail.com>",

package/resolve-advanced/server/author-interchange.mjs

@@ -436,6 +436,7 @@ export function eventsToAssembleSpec(events, opts = {}) {
   const audioTrackNum = (t) => { const m = /^A(\d+)?$/.exec(String(t)); return m && m[1] ? parseInt(m[1], 10) : 1; };
   const audioPlacements = [];
   const audioRetimesSkipped = [];
+  const audioLanesBeyondCeiling = [];
   const audioTransCandidates = [];
   let audioBlackLegsSkipped = 0;
   for (const e of auds) {
@@ -460,6 +461,14 @@ export function eventsToAssembleSpec(events, opts = {}) {
       audioRetimesSkipped.push({ index: e.index, source: e.source, speed: e.speed, reverse: !!e.reverse, reason: 'audio retime not authorable — the audio engine ignores the clip timemap (measured: reads back retimed, renders 100%); played at 100%' });
     }
     const track = audioTrackNum(e.track);
+    // AUDIO LANE CEILING (E135): Resolve authors audio tracks A1–A16 (A9–A16
+    // render-verified); flattened nested sequences can stack lanes far past
+    // that (A40 measured on a real Premiere reels project). Anything above
+    // 16 drops with a reason — named in the ledger, never authored blind.
+    if (track > 16) {
+      audioLanesBeyondCeiling.push({ index: e.index, source: e.source, track: e.track, recIn: e.recIn, recOut: e.recOut, reason: `audio lane ${track} exceeds the A16 ceiling Resolve authors — flatten fewer stacked nested sequences or remix the bed` });
+      continue;
+    }
     const cut = { startFrame: recIn, durationFrames, srcIn: toTl(e.srcIn ?? 0, e.fps), audioOnly: true, track };
     if (e.transition) {
       let d = Math.max(2, toTl(e.transition.duration || 0, e.fps) || 2);
@@ -730,6 +739,7 @@ export function eventsToAssembleSpec(events, opts = {}) {
       unresolvedCompounds,
       authoredAudioEvents: audioPlacements.length,
       audioRetimesSkipped,
+      audioLanesBeyondCeiling,
       ...(elements.length || audioBlackLegsSkipped ? {
         blackLegs: { authoredGenerators: elements.length, audioSilenceLegsSkipped: audioBlackLegsSkipped },
       } : {}),

package/resolve-advanced/server/prproj.mjs

@@ -260,10 +260,17 @@ function walkSequence(seqEntry, byId, depth = 0, seen = new Set()) {
     const refs = asArray(grp.node.TrackGroup?.Tracks?.Track);
     if (refs.length) trackLists.push({ trackKind, refs });
   }
+  // Lanes number per kind in track order — V, V2, V3 … / A, A2 … — like
+  // every other parser (E109/E135): labelling every track 'V'/'A' stacked
+  // 687 of 741 real sequences' cuts onto one lane (4631 overlapping pairs).
+  const laneCount = { V: 0, A: 0 };
   for (const { trackKind, refs: trackRefs } of trackLists) {
     for (const tref of trackRefs) {
       const tEntry = byId.get(refId(tref));
       if (!tEntry) continue;
+      laneCount[trackKind] += 1;
+      const laneNum = laneCount[trackKind];
+      const trackLabel = laneNum === 1 ? trackKind : `${trackKind}${laneNum}`;
       // Real Premiere 2025 keeps a track's TRANSITIONS in a separate list,
       // ClipTrack.TransitionItems (E134, measured: 2 video + 1 audio on the
       // reel — none reached the parser from ClipItems alone).
@@ -279,29 +286,19 @@ function walkSequence(seqEntry, byId, depth = 0, seen = new Set()) {
         const cEntry = byId.get(refId(iref));
         if (!cEntry) continue;
         if (/ClipTrackItem$/.test(cEntry.tag)) {
-          const ev = clipEvent(cEntry.node, byId, trackKind, fps, idx++);
+          const ev = clipEvent(cEntry.node, byId, trackLabel, fps, idx++);
           const nested = ev.__nested;
           delete ev.__nested;
           const nestedKey = nested ? String(nested.node['@_ObjectUID'] ?? nested.node['@_ObjectID'] ?? '') : '';
           if (nested && depth < 8 && !seenHere.has(nestedKey)) {
-            // FLATTEN the nested sequence into this record span (E133, the
-            // OTIO Stack / AAF nested-composition flatten for Premiere): walk
-            // it with its own cursor, keep this track kind's lanes, translate
-            // through the window [srcIn, srcIn + record length), tag fromCompound.
-            const inner = walkSequence(nested, byId, depth + 1, seenHere).filter((e) => e.track.charAt(0) === trackKind && e.recIn != null && e.recOut != null);
-            const winStart = ev.srcIn ?? 0;
-            const winLen = (ev.recOut ?? 0) - (ev.recIn ?? 0);
-            let first = true;
-            for (const e of inner) {
-              const a = e.recIn - winStart, b = e.recOut - winStart;
-              const a2 = Math.max(0, a), b2 = Math.min(winLen, b);
-              if (b2 <= a2) continue;
-              const k = ((e.speed ?? 100) / 100) * (e.reverse ? -1 : 1);
-              const flat = { ...e, index: idx++, recIn: ev.recIn + a2, recOut: ev.recIn + b2, srcIn: e.srcIn != null ? Math.round(e.srcIn + (a2 - a) * k) : e.srcIn, srcOut: e.srcOut != null ? Math.round(e.srcOut - (b - b2) * k) : e.srcOut, fromCompound: ev.source };
-              if (first && ev.transition && !flat.transition) flat.transition = ev.transition;
-              first = false;
-              clipEvents.push(flat);
-            }
+            // Defer the FLATTEN (E133/E135): expanded after every parent lane
+            // is known, so a nested block's inner lanes can first-fit above
+            // whatever the parent already stacks there — two nested
+            // sequences on different parent lanes each bring their own inner
+            // lanes, and a fixed offset collides (measured: 7615 overlapping
+            // pairs on the reels project).
+            ev.__nestedExpand = { nested, laneNum, trackKind, seenHere };
+            clipEvents.push(ev);
             continue;
           }
           if (nested) ev.compound = ev.source; // cycle/depth guard: keep it as a named compound hole
@@ -333,36 +330,120 @@ function walkSequence(seqEntry, byId, depth = 0, seen = new Set()) {
           incoming.transition = trans;
           const outgoing = clipEvents.find((e) => e !== incoming && e.recOut > tr.recIn - 1 && e.recOut <= end);
           if (!outgoing) {
-            clipEvents.push({ index: idx++, track: trackKind, source: 'BL', srcIn: 0, srcOut: 0, recIn: incoming.recIn, recOut: incoming.recIn, speed: 100, reverse: false, transition: null, fps });
+            clipEvents.push({ index: idx++, track: trackLabel, source: 'BL', srcIn: 0, srcOut: 0, recIn: incoming.recIn, recOut: incoming.recIn, speed: 100, reverse: false, transition: null, fps });
           }
         } else {
           const outgoing = clipEvents.find((e) => e.recOut > tr.recIn - 1 && e.recOut <= end);
           if (outgoing) {
-            clipEvents.push({ index: idx++, track: trackKind, source: 'BL', srcIn: 0, srcOut: 0, recIn: outgoing.recOut, recOut: outgoing.recOut, speed: 100, reverse: false, transition: trans, fps });
+            clipEvents.push({ index: idx++, track: trackLabel, source: 'BL', srcIn: 0, srcOut: 0, recIn: outgoing.recOut, recOut: outgoing.recOut, speed: 100, reverse: false, transition: trans, fps });
           }
         }
       }
       events.push(...clipEvents);
     }
   }
-  return events;
+  return expandNestedPlaceholders(events, byId, fps, depth);
+}
+
+/**
+ * Flatten every deferred nested-sequence placeholder into the parent record
+ * time (E133/E135). Inner lanes land on the parent's lane and the lanes above
+ * it, shifted up by the smallest offset at which none of the block's cuts
+ * overlap what the parent (or an earlier block) already holds on those lanes.
+ */
+function expandNestedPlaceholders(events, byId, fps, depth) {
+  const laneOf = (t) => parseInt(String(t).replace(/\D/g, '') || '1', 10);
+  const label = (kind, n) => (n === 1 ? kind : `${kind}${n}`);
+  const occupancy = new Map(); // lane label → [[recIn, recOut], …]
+  const occupy = (e) => { if (e.recIn == null || e.recOut == null) return; (occupancy.get(e.track) || occupancy.set(e.track, []).get(e.track)).push([e.recIn, e.recOut]); };
+  const free = (lane, a, b) => !(occupancy.get(lane) || []).some(([x, y]) => a < y && b > x);
+  for (const e of events) if (!e.__nestedExpand) occupy(e);
+  const out = [];
+  let idx = 1;
+  for (const ev of events) {
+    const ph = ev.__nestedExpand;
+    if (!ph) { out.push({ ...ev, index: idx++ }); continue; }
+    delete ev.__nestedExpand;
+    const { nested, laneNum, trackKind, seenHere } = ph;
+    const inner = walkSequence(nested, byId, depth + 1, seenHere).filter((e) => e.track.charAt(0) === trackKind && e.recIn != null && e.recOut != null);
+    const winStart = ev.srcIn ?? 0;
+    const winLen = (ev.recOut ?? 0) - (ev.recIn ?? 0);
+    const block = [];
+    for (const e of inner) {
+      const a = e.recIn - winStart, b = e.recOut - winStart;
+      const a2 = Math.max(0, a), b2 = Math.min(winLen, b);
+      if (b2 <= a2) continue;
+      const k = ((e.speed ?? 100) / 100) * (e.reverse ? -1 : 1);
+      block.push({ ...e, recIn: ev.recIn + a2, recOut: ev.recIn + b2, srcIn: e.srcIn != null ? Math.round(e.srcIn + (a2 - a) * k) : e.srcIn, srcOut: e.srcOut != null ? Math.round(e.srcOut - (b - b2) * k) : e.srcOut, fromCompound: ev.source, __innerLane: laneOf(e.track) });
+    }
+    // First-fit: the smallest upward shift at which every cut of the block finds its lane free.
+    let shift = 0;
+    for (; shift < 64; shift += 1) {
+      if (block.every((e) => free(label(trackKind, laneNum + e.__innerLane - 1 + shift), e.recIn, e.recOut))) break;
+    }
+    let first = true;
+    for (const e of block) {
+      const placed = { ...e, index: idx++, track: label(trackKind, laneNum + e.__innerLane - 1 + shift) };
+      delete placed.__innerLane;
+      if (shift) placed.laneShift = shift;
+      if (first && ev.transition && !placed.transition) placed.transition = ev.transition;
+      first = false;
+      occupy(placed);
+      out.push(placed);
+    }
+    if (!block.length) out.push({ ...ev, index: idx++, compound: ev.source });
+  }
+  return out;
 }
 
-/** Enumerate marker objects (project- or sequence-level) with tick→frame positions. */
-function collectMarkers(byId, fps) {
+/**
+ * A sequence's OWN markers (E136). Real Premiere 2025: Sequence.MarkerOwner.
+ * Markers → a Markers container → <Markers><Marker><Second ObjectRef> → a
+ * Marker object whose payload is a DVAMarker JSON string ({mStartTime.ticks,
+ * mName, mComment, mType, mMarkerID}; a duration/end when present). The old
+ * walk returned EVERY Marker object in the project on EVERY sequence, with
+ * frame 0 and an empty name (1228 bogus markers per sequence on a real
+ * turnover). Legacy/synthetic Marker objects (Position/Name/Comment fields)
+ * are read the same way when a sequence owns them; without an owner chain a
+ * sequence reports none.
+ */
+function markerFromObject(node, fps) {
+  const raw = childText(node, 'DVAMarker');
+  if (raw) {
+    try {
+      const m = JSON.parse(String(raw)).DVAMarker || {};
+      const startTicks = m.mStartTime && m.mStartTime.ticks != null ? Number(m.mStartTime.ticks) : null;
+      const endTicks = m.mEndTime && m.mEndTime.ticks != null ? Number(m.mEndTime.ticks) : null;
+      const frame = startTicks != null ? ticksToFrames(startTicks, fps) : null;
+      const duration = startTicks != null && endTicks != null ? Math.max(0, ticksToFrames(endTicks - startTicks, fps) || 0) : 0;
+      return { frame, duration, name: m.mName || '', note: m.mComment || '', type: m.mType || 'Comment', id: m.mMarkerID || null };
+    } catch {
+      return null;
+    }
+  }
+  return {
+    frame: ticksToFrames(childText(node, 'Position'), fps),
+    duration: ticksToFrames(childText(node, 'Duration'), fps) || 0,
+    name: childText(node, 'Name') || '',
+    note: childText(node, 'Comment') || '',
+    type: Number(childText(node, 'MarkerType')) || 0,
+    colorIndex: Number(childText(node, 'ColorIndex')) || 0,
+  };
+}
+
+function collectMarkers(seqEntry, byId, fps) {
+  const owner = asArray(seqEntry.node.MarkerOwner)[0];
+  const container = owner ? byId.get(ref(owner, 'Markers') || '') : null;
+  if (!container) return [];
+  const list = asArray(container.node.Markers && container.node.Markers.Marker);
   const markers = [];
-  for (const { tag, node } of byId.values()) {
-    if (tag !== 'Marker') continue;
-    markers.push({
-      frame: ticksToFrames(childText(node, 'Position'), fps),
-      duration: ticksToFrames(childText(node, 'Duration'), fps) || 0,
-      name: childText(node, 'Name') || '',
-      note: childText(node, 'Comment') || '',
-      type: Number(childText(node, 'MarkerType')) || 0,
-      colorIndex: Number(childText(node, 'ColorIndex')) || 0,
-    });
+  for (const pair of list) {
+    const mk = byId.get(refId(asArray(pair.Second)[0]) || '');
+    if (!mk || mk.tag !== 'Marker') continue;
+    const m = markerFromObject(mk.node, fps);
+    if (m && m.frame != null) markers.push(m);
   }
-  return markers;
+  return markers.sort((a, b) => a.frame - b.frame);
 }
 
 function sequenceEntries(byId) {
@@ -387,7 +468,7 @@ export function parsePrprojDoc(pathOrBuffer) {
       fps,
       eventCount: events.length,
       events,
-      markers: collectMarkers(byId, fps),
+      markers: collectMarkers(entry, byId, fps),
     };
   });
   const mediaPaths = [...new Set(collectMediaPaths(byId))].sort();

package/resolve-advanced/server/tools/drt.mjs

@@ -154,7 +154,7 @@ function requirePathArg(args, key, action) {
 export const drtTool = {
   name: 'drt',
   description:
-    'DaVinci Resolve Timeline (.drt) operations — offline, no Resolve required. Actions: assemble_from_interchange (EDL/OTIO/XML/AAF + sourceMap → IMPORTABLE RENDERING native .drt in one call; retimes AUTHOR — constant speed fwd/rev AND zero-speed freezes (EDL M2 000.0; render-proven frozen); cross-dissolves are AUTHORED when the cut abuts with handles both sides (render-verified on 19), else dropped with reason; fades AUTHOR across ALL FOUR formats (EDL BL legs, OTIO gap-adjacent Transitions, XMEML edge transitionitems, AAF filler-adjacent Transitions) and AAF overlap-consuming dissolves reconcile+author (CutPoint honored; before E93 they threw at the overlap gate) — BL legs become Solid Color generators and the fade a real clip-to-generator dissolve, luma-ramp render-verified, while Resolve\'s OWN EDL importer drops BL dissolves silently; ledger in `conform`), assemble (spec → IMPORTABLE native-schema .drt via template-spliced real structures; pass targetAppVersion e.g. \'19.1\' for pre-21 hosts), parse, list_sequences (enumerate the timelines inside a .drp/.drt → [{id,name,eventCount,index,kind,nestedIn}] to drive a "which sequence?" picker — names and kind (timeline|compound) come from the pool folder, since a SeqContainer carries no timeline name and its first <Name> is a CLIP\'s (measured, E127); a compound container is nestedIn every timeline that places it, and a media-less clip named after a compound is tagged `compound` in parse), author, validate, inject_into_drp, extract_from_drp (pull one SeqContainer out as a .drt — by timelineName, an explicit timelineIndex, or by default the first container the pool kinds as a TIMELINE (index 0 is name-sorted by DbId and was an inner compound on Resolve\'s own export, E128); a timeline keeps its compound containers recursively; feed the .drt to the Python davinci-resolve MCP timeline.import_timeline_checked, or use timeline.import_from_drp to do both), downgrade (stamp <ProjectVersion> down so an OLDER Resolve will import a .drt/.drp from a newer one — pass targetAppVersion like "19.1.3" or targetProjectVersion).',
+    'DaVinci Resolve Timeline (.drt) operations — offline, no Resolve required. Actions: assemble_from_interchange (EDL/OTIO/XML/AAF + sourceMap → IMPORTABLE RENDERING native .drt in one call; retimes AUTHOR — constant speed fwd/rev AND zero-speed freezes (EDL M2 000.0; render-proven frozen); cross-dissolves are AUTHORED when the cut abuts with handles both sides (render-verified on 19), else dropped with reason; audio events above lane A16 drop with a reason in audioLanesBeyondCeiling (flattened nested sequences can stack lanes to A40 — measured); fades AUTHOR across ALL FOUR formats (EDL BL legs, OTIO gap-adjacent Transitions, XMEML edge transitionitems, AAF filler-adjacent Transitions) and AAF overlap-consuming dissolves reconcile+author (CutPoint honored; before E93 they threw at the overlap gate) — BL legs become Solid Color generators and the fade a real clip-to-generator dissolve, luma-ramp render-verified, while Resolve\'s OWN EDL importer drops BL dissolves silently; ledger in `conform`), assemble (spec → IMPORTABLE native-schema .drt via template-spliced real structures; pass targetAppVersion e.g. \'19.1\' for pre-21 hosts), parse, list_sequences (enumerate the timelines inside a .drp/.drt → [{id,name,eventCount,index,kind,nestedIn}] to drive a "which sequence?" picker — names and kind (timeline|compound) come from the pool folder, since a SeqContainer carries no timeline name and its first <Name> is a CLIP\'s (measured, E127); a compound container is nestedIn every timeline that places it, and a media-less clip named after a compound is tagged `compound` in parse), author, validate, inject_into_drp, extract_from_drp (pull one SeqContainer out as a .drt — by timelineName, an explicit timelineIndex, or by default the first container the pool kinds as a TIMELINE (index 0 is name-sorted by DbId and was an inner compound on Resolve\'s own export, E128); a timeline keeps its compound containers recursively; feed the .drt to the Python davinci-resolve MCP timeline.import_timeline_checked, or use timeline.import_from_drp to do both), downgrade (stamp <ProjectVersion> down so an OLDER Resolve will import a .drt/.drp from a newer one — pass targetAppVersion like "19.1.3" or targetProjectVersion).',
   async handler({ action, args }) {
     if (action === 'parse') {
       const p = parseSchema.parse(requirePathArg(args, 'drtPath', 'parse'));

package/resolve-advanced/server/tools/editorial.mjs

@@ -86,7 +86,7 @@ const markerSchema = z.object({
 export const editorialTool = {
   name: 'editorial',
   description:
-    'Editorial integrity (Cluster E) — turnover interchange → normalized events → changelist + conform manifest with TIMING silent-lie guards (flattened retime / dropped J/L-cut audio / framerate-pulldown slip / reverse dropped / transition-handle starvation → flag, skip-not-fake). Report-only (gate: review). Actions: parse_interchange (EDL/OTIO/XMEML natively — incl. Resolve-written OTIO generator clips (a Solid Color is a Clip with a NULL media_reference; it walks as a BL leg with generatorName, and an OTIO GeneratorReference carries its colour) and nested OTIO Stacks (compound clips — Resolve\'s writer nests them with a trim window) FLATTEN into record time with fromCompound on each cut, so Resolve\'s own OTIO exports re-conform (its FCP7 XML writer flattens a compound to ONE media-less clipitem instead — tagged `compound`, which the bridge drops with a reason in unresolvedCompounds unless the sourceMap maps the compound name to a flattened media file), Resolve-written FCP7 -1 junction edges (paired in record order) + Solid Color / Color Matte generators carrying their `fillcolor` (BL legs with `color` → fade-to-white and colour mattes author, E110), XMEML audio-track transitionitems as audio cross-fades on numbered lanes A/A2/…, and CMX FROM/TO CLIP NAME comments over the generic AX reel — + AAF via pyaaf2 (flat sound/picture slots number A/A2/… and V/V2/… in slot order so separate beds keep their lanes; NestedScope layers keep layer numbering; a NESTED SEQUENCE used as a clip — a SourceClip referencing a named CompositionMob — flattens through its reference window with fromCompound, like OTIO Stacks) + PRPROJ via gunzip+XML (real Premiere 2025 files — uuid ObjectUID/ObjectURef graph, TrackGroups, ClipTrackItem→SubClip→VideoClip→MediaSource→Media chain, zeros written as absence — and the legacy ObjectID/VideoTracks shape both walk; measured on a 739-sequence turnover that used to list ZERO; a NESTED SEQUENCE used as a clip — Source = Video/AudioSequenceSource → Sequence — flattens through its in-point window with fromCompound, like OTIO Stacks and AAF nested compositions; 13,711 such events on one real reels project; real transitions live in the track\'s TransitionItems list — read alongside ClipItems, attached with their DisplayName as type, HasOutgoingClip/HasIncomingClip false = fade from/to black or silence) → normalized events incl. span-explicit transitions, BL fade legs, and freezes as zero-speed events (OTIO FreezeFrame, XMEML timeremap 0, PrProj in==out, AAF 0% motion effects); for AAF/PRPROJ pass the file PATH as content; AAF also returns per-sequence startTimecode/startFrame — build the timeline at THAT start, not the Resolve 01:00:00:00 default — and per-clip `geometry` for Avid transform effects), list_sequences (ONE offline picker entry point across xml/edl/otio/drt/drp/aaf/prproj → [{id,name,eventCount}], plus startTimecode/startFrame for AAF and nestedIn — an AAF composition another composition uses as a clip is NESTED in it; offer the parent, its cuts arrive flattened), convert_to_interchange (author OTIO/EDL/DRT Resolve CAN import from events or a parsed source; the EDL target writes CMX transition pairs incl. BL fades — the .prproj→Resolve conform bridge, no Premiere needed; editorial timing/transitions survive and per-clip effects/color do not. SPEED/REVERSE survive on the otio (LinearTimeWarp) and edl (M2) targets ONLY — this FLAT drt target flattens every retime to 100% forward and returns `flattened`/`flattenedCount` naming each event that lost one (`flattened` is always present on `drt`, empty when there were none); for a .drt that AUTHORS retimes/dissolves/multi-track/audio, use drt.assemble_from_interchange), turnover_changelist (diff old vs new → moved/retimed/trimmed/replaced/new/gone PLUS the junction diff: transition_added/transition_dropped/transition_changed with fade in/out or dissolve, outgoing/incoming, span and duration/type/pre-roll deltas — zero-length CMX carrier lines and the BL legs that carry fades fold into the junction diff instead of reading as gone/new sources; events pair by closest record position, consumed once, so a source cut twice at two speeds compares instance to instance; + timing flags incl. transition_dropped and dropped_split_audio on any A-track; a compound seen collapsed in one cut and flattened in the other reports compound_collapsed/compound_expanded once, never replaced+gone), conform_manifest (per-event assert: source resolved/handles/retime/reverse/TC-base; BL-aware — black legs need no source, fades no black-side handles, and a fade-out tail requirement lands on the picture source; a compound clipitem fails source_resolved by NAME with the remedy — map it to a flattened file or turn over as OTIO), marker_roundtrip (markers with provenance tags), verify_roundtrip (input events vs re-export events -> pass/mismatches + fitted per-source TC offsets + marker compare w/ markersNotInExport honesty flag; FADE-AWARE: BL/Solid-Color legs merge out as blackSegments and fade-window boundary reshapes are excused into fadeReshapedBoundaries instead of failing; RETIME-AWARE: speed/reverse compare pairwise — EXPORT_OTIO carries an authored Sm2TimeMap back as LinearTimeWarp (measured), so a flattened/lost retime fails as drift geometry alone cannot catch; AUDIO-AWARE: declared audio events compare (a video-only export such as EXPORT_EDL flags audioNotInExport instead of failing) (channel legs deduped, mismatches tagged trackType audio) while the mirrored-A1 export of a video-only turnover stays informational; COLOUR-AWARE: an input generator leg carrying a fillcolor (fade-to-white, colour matte) must come back on the same track over its span with the same colour — Resolve\'s FCP7 writer emits it — else generator-colour fails (generatorColours reports the compare) — pass exportedFormat: an OTIO/EDL re-export cannot carry colour (measured) and reports generatorColourNotInExport instead of failing; COMPOUND-AWARE: an XML re-export that collapsed a compound to one clipitem over cuts the input flattened from that compound reports compoundsCollapsedInExport instead of count/source drift; the conform QC loop-closer). Offline (AAF needs pyaaf2; live AAF/DRP import is on the Python davinci-resolve MCP).',
+    'Editorial integrity (Cluster E) — turnover interchange → normalized events → changelist + conform manifest with TIMING silent-lie guards (flattened retime / dropped J/L-cut audio / framerate-pulldown slip / reverse dropped / transition-handle starvation → flag, skip-not-fake). Report-only (gate: review). Actions: parse_interchange (EDL/OTIO/XMEML natively — incl. Resolve-written OTIO generator clips (a Solid Color is a Clip with a NULL media_reference; it walks as a BL leg with generatorName, and an OTIO GeneratorReference carries its colour) and nested OTIO Stacks (compound clips — Resolve\'s writer nests them with a trim window) FLATTEN into record time with fromCompound on each cut, so Resolve\'s own OTIO exports re-conform (its FCP7 XML writer flattens a compound to ONE media-less clipitem instead — tagged `compound`, which the bridge drops with a reason in unresolvedCompounds unless the sourceMap maps the compound name to a flattened media file), Resolve-written FCP7 -1 junction edges (paired in record order) + Solid Color / Color Matte generators carrying their `fillcolor` (BL legs with `color` → fade-to-white and colour mattes author, E110), XMEML audio-track transitionitems as audio cross-fades on numbered lanes A/A2/…, and CMX FROM/TO CLIP NAME comments over the generic AX reel — + AAF via pyaaf2 (flat sound/picture slots number A/A2/… and V/V2/… in slot order so separate beds keep their lanes; NestedScope layers keep layer numbering; a NESTED SEQUENCE used as a clip — a SourceClip referencing a named CompositionMob — flattens through its reference window with fromCompound, like OTIO Stacks) + PRPROJ via gunzip+XML (real Premiere 2025 files — uuid ObjectUID/ObjectURef graph, TrackGroups, ClipTrackItem→SubClip→VideoClip→MediaSource→Media chain, zeros written as absence — and the legacy ObjectID/VideoTracks shape both walk; measured on a 739-sequence turnover that used to list ZERO; a NESTED SEQUENCE used as a clip — Source = Video/AudioSequenceSource → Sequence — flattens through its in-point window with fromCompound, like OTIO Stacks and AAF nested compositions; 13,711 such events on one real reels project; real transitions live in the track\'s TransitionItems list — read alongside ClipItems, attached with their DisplayName as type, HasOutgoingClip/HasIncomingClip false = fade from/to black or silence; tracks number as lanes V/V2…/A/A2… and a flattened nested sequence\'s inner lanes first-fit above what the parent already stacks — laneShift records the shift; markers are the sequence\'s OWN, read from Premiere\'s DVAMarker JSON) → normalized events incl. span-explicit transitions, BL fade legs, and freezes as zero-speed events (OTIO FreezeFrame, XMEML timeremap 0, PrProj in==out, AAF 0% motion effects); for AAF/PRPROJ pass the file PATH as content; AAF also returns per-sequence startTimecode/startFrame — build the timeline at THAT start, not the Resolve 01:00:00:00 default — and per-clip `geometry` for Avid transform effects), list_sequences (ONE offline picker entry point across xml/edl/otio/drt/drp/aaf/prproj → [{id,name,eventCount}], plus startTimecode/startFrame for AAF and nestedIn — an AAF composition another composition uses as a clip is NESTED in it; offer the parent, its cuts arrive flattened), convert_to_interchange (author OTIO/EDL/DRT Resolve CAN import from events or a parsed source; the EDL target writes CMX transition pairs incl. BL fades — the .prproj→Resolve conform bridge, no Premiere needed; editorial timing/transitions survive and per-clip effects/color do not. SPEED/REVERSE survive on the otio (LinearTimeWarp) and edl (M2) targets ONLY — this FLAT drt target flattens every retime to 100% forward and returns `flattened`/`flattenedCount` naming each event that lost one (`flattened` is always present on `drt`, empty when there were none); for a .drt that AUTHORS retimes/dissolves/multi-track/audio, use drt.assemble_from_interchange), turnover_changelist (diff old vs new → moved/retimed/trimmed/replaced/new/gone PLUS the junction diff: transition_added/transition_dropped/transition_changed with fade in/out or dissolve, outgoing/incoming, span and duration/type/pre-roll deltas — zero-length CMX carrier lines and the BL legs that carry fades fold into the junction diff instead of reading as gone/new sources; events pair by closest record position, consumed once, so a source cut twice at two speeds compares instance to instance; + timing flags incl. transition_dropped and dropped_split_audio on any A-track; a compound seen collapsed in one cut and flattened in the other reports compound_collapsed/compound_expanded once, never replaced+gone), conform_manifest (per-event assert: source resolved/handles/retime/reverse/TC-base; BL-aware — black legs need no source, fades no black-side handles, and a fade-out tail requirement lands on the picture source; a compound clipitem fails source_resolved by NAME with the remedy — map it to a flattened file or turn over as OTIO), marker_roundtrip (markers with provenance tags), verify_roundtrip (input events vs re-export events -> pass/mismatches + fitted per-source TC offsets + marker compare w/ markersNotInExport honesty flag; FADE-AWARE: BL/Solid-Color legs merge out as blackSegments and fade-window boundary reshapes are excused into fadeReshapedBoundaries instead of failing; RETIME-AWARE: speed/reverse compare pairwise — EXPORT_OTIO carries an authored Sm2TimeMap back as LinearTimeWarp (measured), so a flattened/lost retime fails as drift geometry alone cannot catch; AUDIO-AWARE: declared audio events compare (a video-only export such as EXPORT_EDL flags audioNotInExport instead of failing) (channel legs deduped, mismatches tagged trackType audio) while the mirrored-A1 export of a video-only turnover stays informational; COLOUR-AWARE: an input generator leg carrying a fillcolor (fade-to-white, colour matte) must come back on the same track over its span with the same colour — Resolve\'s FCP7 writer emits it — else generator-colour fails (generatorColours reports the compare) — pass exportedFormat: an OTIO/EDL re-export cannot carry colour (measured) and reports generatorColourNotInExport instead of failing; COMPOUND-AWARE: an XML re-export that collapsed a compound to one clipitem over cuts the input flattened from that compound reports compoundsCollapsedInExport instead of count/source drift; the conform QC loop-closer). Offline (AAF needs pyaaf2; live AAF/DRP import is on the Python davinci-resolve MCP).',
   async handler({ action, args }) {
     if (action === 'parse_interchange') {
       const p = parseSchema.parse(args);

package/src/granular/common.py

@@ -87,7 +87,7 @@ if not logging.getLogger().handlers:
         handlers=[logging.StreamHandler()],
     )
 
-VERSION = "2.187.0"
+VERSION = "2.189.0"
 logger = logging.getLogger("davinci-resolve-mcp")
 logger.info(f"Starting DaVinci Resolve MCP Server v{VERSION}")
 logger.info(f"Detected platform: {get_platform()}")

package/src/server.py

@@ -11,7 +11,7 @@ Usage:
     python src/server.py --full       # Start the 353-tool granular server instead
 """
 
-VERSION = "2.187.0"
+VERSION = "2.189.0"
 
 import base64
 import os

package/src/utils/api_truth.py

@@ -2919,7 +2919,14 @@ API_TRUTH: List[Dict[str, Any]] = [
                        "TransitionTrackItem.TrackItem{Start,End} (the span), "
                        "DisplayName / MatchName (the effect), Alignment (ticks) "
                        "and HasIncomingClip / HasOutgoingClip — false marks a "
-                       "fade from/to black or silence.",
+                       "fade from/to black or silence. MARKERS (E136): a "
+                       "sequence owns its markers through Sequence.MarkerOwner."
+                       "Markers → a Markers container → <Markers><Marker><Second "
+                       "ObjectRef> → a Marker object whose payload is a DVAMarker "
+                       "JSON string ({mMarkerID, mStartTime.ticks, mType, mName?, "
+                       "mComment?, mEndTime?}); clips own theirs the same way via "
+                       "Clip.MarkerOwner. Marker objects exist project-wide (1228 "
+                       "on the turnover) — only the owner chain says whose they are.",
         "tags": ["prproj", "premiere", "interchange", "silent-failure"],
         "submit": "missing",
         "mitigation": ["editorial.parse_interchange prproj", "editorial.list_sequences"],