v0.5 roadmap — vote on what ships next

[renezander030]

Help shape v0.5. capcut-cli grew fast this past month (v0.3 + v0.4 shipped decorators, SRT import, Wikimedia, schema docs, version resilience, caption factory — see the changelog). For v0.5 I want to ship what people actually use, not what looks shiny.

How to vote: 👍 react to the comment of the feature you want most. Multi-vote welcome. If your feature is missing, drop a new comment and I'll move it into the list.

The current candidate set (each gets its own comment below):

1. audio-fade — fade-in / fade-out on audio segments (cleaner than volume keyframes; pyJianYingDraft re-shipped this in March, recurring demand)
2. bubble-text / 花字 — bubble/decorative text effects (30+ comments across 3 pyJianYingDraft issues — single biggest request)
3. add-filter + enums --filters — colour filter chain (scene effects exist, colour filters don't — capcut-mate has this)
4. add-cover / 封面 — set the cover frame from the shell (no tool in the ecosystem solves this cleanly)
5. import-ass — ASS subtitle import alongside SRT (anime / CN fansub format, requested in pyJianYingDraft #64)
6. mix-mode — blend modes per video segment (pyJianYingDraft shipped this in March 2026; trivial JSON-field port)

I'll ship the top 3-5 by reaction count as a single v0.5 release. Voting closes when v0.5 is ready (target: 2 weeks).

If you're using capcut-cli in production / a pipeline, please mention what for in a comment — context shapes scope.

[renezander030]

audio-fade <project> <id> --in <s> --out <s> — proper fade objects on the audio material's audio_fades field (not volume keyframes). pyJianYingDraft shipped this Sept 2025 and re-shipped March 2026 → proves recurring real-world demand. ~half-day effort.

[renezander030]

bubble-text <project> <id> --bubble <slug> + enums --bubbles for discovery. Sets bubble_* fields on the text material; --jianying flag for the JianYing enum namespace. The single biggest request in the pyJianYingDraft tracker (issues #70 + #44 + #132, 30+ comments). Critical for Douyin/Xiaohongshu title styles.

[renezander030]

add-filter <project> <slug> <start> <duration> + enums --filters. Colour filter chain — currently add-effect only handles VFX/scene effects. capcut-mate ships this; we lose a comparison-table row without it. Mirrors add-effect impl.

[renezander030]

add-cover <project> <image-path> [--time <ms>] — sets the JianYing/CapCut cover frame from the shell. Nobody in the competitor list solves this cleanly (issue #56 in pyJianYingDraft). Pure JSON write on cover.{image,time}. Quick win, unique-to-us positioning.

[renezander030]

import-ass <project> <ass-path> alongside existing import-srt. ASS subtitle format support — big for anime/CN fansub workflows. Zero-dep ASS parser ~150 lines. Reuses --style-ref path from import-srt. Requested as pyJianYingDraft #64.

[renezander030]

mix-mode <project> <id> <mode> — blend modes per video segment (multiply, screen, overlay, etc.). pyJianYingDraft shipped this in March 2026 — currently hot in CN community. Single JSON-field write on video.material. Ships in same PR as audio-fade if convenient.

[renezander030]

v0.5 shipped — thanks for voting.

All six candidate features made it into v0.5.0:

• capcut mix-mode — 12 blend modes on video segments
• capcut audio-fade --in / --fade-out — real materials.audio_fades[] entry
• capcut add-cover — set the draft cover frame from the CLI
• capcut add-filter + capcut enums --filters — colour-filter track separate from add-effect
• capcut bubble-text + capcut enums --bubbles — speech-bubble shape on a text segment
• capcut import-ass — ASS / SSA subtitle import alongside import-srt

Released as v0.5.0 today: https://github.com/renezander030/capcut-cli/releases/tag/v0.5.0

Install: npm install -g capcut-cli@0.5.0

If anything misbehaves on your draft schema (CapCut version diffs especially around the cover and bubble field shapes), please open an issue with the capcut version output and a minimal repro.