This release includes breaking changes for platform teams planning a safe upgrade.
✓ No known CVEs patched in this version
Topics
+14 more
ReleasePort's take
Light signalRelease v1.5.0 hardens the explainer workflow into a 10‑stage lane with audio normalization, transition audit records, quality reporting, and enforces a min_score of 50.
Why it matters: The new 10‑stage pipeline introduces a mandatory min_score=50 checkpoint, directly affecting any process that relies on the explainer workflow; operators must verify compliance before deployment.
Summary
AI summaryHardened the explainer workflow with a 10‑stage lane, audio normalization, transition audit records, quality report, and a real min_score=50 release checkpoint.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Medium |
Adds Video Receipts as review trail for agent media. Adds Video Receipts as review trail for agent media. Source: llm_adapter@2026-06-04 Confidence: high |
— |
| Feature | Medium |
Adds receipt-backed confidence baseline and local repurpose-package workflows. Adds receipt-backed confidence baseline and local repurpose-package workflows. Source: llm_adapter@2026-06-04 Confidence: high |
— |
| Feature | Medium |
Adds confidence benchmark and adversarial certification scripts for local readiness proof. Adds confidence benchmark and adversarial certification scripts for local readiness proof. Source: llm_adapter@2026-06-04 Confidence: high |
— |
| Dependency | Low |
Removes stale `/explainer-video` npm Dependabot entry to enable real dependency automation. Removes stale `/explainer-video` npm Dependabot entry to enable real dependency automation. Source: llm_adapter@2026-06-04 Confidence: high |
— |
| Bugfix | Medium |
Makes corrupted media probing fail clearly as InputFileError. Makes corrupted media probing fail clearly as InputFileError. Source: llm_adapter@2026-06-04 Confidence: high |
— |
| Bugfix | Medium |
Fixes confidence benchmark so stale receipts cannot hide a failed workflow run. Fixes confidence benchmark so stale receipts cannot hide a failed workflow run. Source: llm_adapter@2026-06-04 Confidence: high |
— |
| Refactor | Low |
Hardens explainer workflow into a 10-stage lane with audio normalization, transition audit records, quality report, and min_score=50 checkpoint. Hardens explainer workflow into a 10-stage lane with audio normalization, transition audit records, quality report, and min_score=50 checkpoint. Source: llm_adapter@2026-06-04 Confidence: high |
— |
Full changelog
Highlights
- Added Video Receipts as the review trail for agent-created or agent-edited media.
- Added receipt-backed confidence baseline and local repurpose-package workflows.
- Added confidence benchmark and adversarial certification scripts for local readiness proof.
- Hardened the explainer workflow into a 10-stage lane with audio normalization, transition audit records, quality report, and a real
min_score=50release checkpoint.
Fixes
- Removed the stale
/explainer-videonpm Dependabot entry so dependency automation can run against real package paths. - Made corrupted media probing fail clearly as
InputFileError. - Fixed the confidence benchmark so stale receipts cannot hide a failed workflow run.
Verification
- PR #342 merged with Hosted PR checks, agent-law, and CodeRabbit passing.
- PR #343 proved Dependabot automation is alive after the config fix.
- PR #344 aligned
pyproject.toml,mcp_video.__version__, andserver.jsonat1.5.0. - Local package build and
twine checkpassed formcp_video-1.5.0.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About pastorsimon1798/mcp-video
Video editing MCP server with 26 tools for trimming, merging, text overlays, audio sync, filters, color grading, audio normalization, picture-in-picture, split-screen, batch processing, format conversion, subtitles, watermarks, and more. 380 tests, CI on Python 3.11+3.12, progress callbacks, works with Claude Code, Cursor, and any MCP client.
Related context
Beta — feedback welcome: [email protected]