video-context-mcp-server 0.47.1-beta → 0.47.2-beta

package/README.md

@@ -300,12 +300,12 @@ When an audio key is missing or an audio API call fails at runtime, tools automa
 
 Set `VIDEO_MCP_DEFAULT_PROVIDER=gemini`, `qwen`, or `kimi` to change the default.
 
-> **ℹ️ Large local files with GLM or Qwen:** Both providers have a 10–12 MB base64 limit for local files. When a file exceeds this limit, the server **automatically falls back to frame-based analysis** (extracts evenly-spaced keyframes and sends them as images) — no configuration needed. For the highest quality with large local videos, set up the optional **S3 relay** (below) — GLM and Qwen will receive a presigned URL to the full video instead of extracted frames.
+> **ℹ️ Large local files with GLM or Qwen:** Both providers have a 10–12 MB base64 limit for local files. When a file exceeds this limit, the server first tries to fall back to an upload-capable provider (Gemini or Kimi) if one is available in the fallback chain. Frame-based analysis (evenly-spaced keyframes sent as images) is used only as a **last resort** when no upload-capable provider is available — no configuration needed. For the highest quality with large local videos, set up the optional **S3 relay** (below) — GLM and Qwen will receive a presigned URL to the full video, bypassing the limit entirely and taking priority over both fallbacks.
 
 <details open>
 <summary><strong>Automatic S3 relay: bypass the 10 MB local file limit with Qwen and GLM</strong></summary>
 
-**GLM-4.6V** and **Qwen3.5** both accept direct video URLs, but base64-encoding a local file caps out at **10–12 MB**. Above that limit, the server automatically falls back to **frame-based analysis** (good quality). For the best results on large local videos, set `AWS_S3_BUCKET` — the server uploads the full video to S3 and passes a presigned URL to GLM/Qwen, bypassing the base64 limit entirely. No manual upload step needed.
+**GLM-4.6V** and **Qwen3.5** both accept direct video URLs, but base64-encoding a local file caps out at **10–12 MB**. Above that limit, the server first tries to fall back to an upload-capable provider (Gemini or Kimi) if one is available, then falls back to **frame-based analysis** as a last resort. For the best results on large local videos, set `AWS_S3_BUCKET` — the server uploads the full video to S3 and passes a presigned URL to GLM/Qwen, bypassing the base64 limit entirely and taking priority over both fallbacks. No manual upload step needed.
 
 **Why it works**: Qwen and GLM require the server to serve `Content-Length` and `Content-Type` headers alongside the video. AWS S3 provides these automatically.
 

package/dist/generated/version.d.ts

@@ -1,2 +1,2 @@
-export declare const VERSION = "0.47.1-beta";
+export declare const VERSION = "0.47.2-beta";
 //# sourceMappingURL=version.d.ts.map

package/dist/generated/version.js

@@ -1,3 +1,3 @@
 // Auto-generated by scripts/sync-version.ts — do not edit
-export const VERSION = '0.47.1-beta';
+export const VERSION = '0.47.2-beta';
 //# sourceMappingURL=version.js.map

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "video-context-mcp-server",
-  "version": "0.47.1-beta",
+  "version": "0.47.2-beta",
   "description": "A Model Context Protocol server that gives GitHub Copilot the ability to understand and analyze video content",
   "type": "module",
   "main": "dist/index.js",