video-quality-enhance

Video Quality Enhance

Overview

Call meitu video-quality-enhance for comprehensive video quality restoration: deblurring + denoising + frame interpolation + dark area brightening, with self-adaptation for all scenarios including portraits/concerts/anime/night scenes/products. Single-item quality requests (denoising/frame interpolation/dark area enhancement) are also handled by this tool by default.

Does not accept target_resolution, does not handle resolution/super-resolution/upscaling-related requests, does not handle images.

API Mapping

DAG tool	Backend API
video-quality-enhance	video_hd_enhance

Dependencies

• meitu-cli: >=2.0.6
• Credentials：CONFIG AKSK → meitu tools update; EXEC AKSK → run command
• Environment variables：MEITU_OPENAPI_TOOL_TASK_MODE=command

Path alias:$VISUAL = {OPENCLAW_HOME}/workspace/visual/

Core Workflow

Preflight → Execute → Deliver

Preflight

1. meitu --version ≥ 2.0.6；meitu auth verify --json
2. Confirm Config Phase has been run;MEITU_OPENAPI_TOOL_TASK_MODE=command
3. output_dir resolution: openclaw.yaml → ./output/ | else → $VISUAL/output/video-quality-enhance/；mkdir -p {output_dir}

Execute

Trigger signals/routing

Matched signal words: "quality restoration, video restoration, one-click restoration, overall restoration, comprehensive restoration, make clearer, old video renovation, quality improvement, one-click enhancement".

Decision order (capability scope judgment has the highest priority):

1. Does the input contain resolution/super-resolution requests (720P/1080P/2K/4K / upscaling / super-resolution / resolution)?
   • Yes → This tool does not handle it, return out of capability scope(guide video-resolution-upscale）
   • No → proceed to next step
2. Agent identifies video content type from the footage and generates prompt
3. video_url / prompt missing → ask follow-up
4. Unique route → video_hd_enhance

prompt → internal scenario adaptation

prompt signal words	Internal adaptation
Portrait/selfie/person/face	Portrait restoration
Concert/live/stage/music event	Concert restoration
Anime/animation/ACGN/cartoon	Anime restoration
Night scene/low light/night shot/low light	Night scene restoration
Product/item/display/e-commerce	Product restoration
Other/general	General restoration (default)

Unable to determine → prompt write "general video" to use the default restoration mode.

Parameter definitions

Parameter	Type	Required	Description
video_url	STRING	Yes	Video URL; missing → "Please provide the video link"
prompt	STRING	Yes	Agent's description of the video content, used for internal multi-scenario adaptation

Does not accept target_resolution parameter—this tool does not handle resolution.

Tool invocation

meitu video-quality-enhance \
  --skill_name skill_video-quality-enhance \
  --video_url {video_url} \
  --prompt "{prompt}" \
  --json --download-dir {output_dir}

Error fallback

Scenario	Handling
video_url missing	Prompt "Please provide the video link"
video_url inaccessible	Do not retry, prompt "Please provide a valid video link"
prompt empty or no signal words	Generated by the Agent after identification; if still unable to determine, write "general video"
Invocation failed	Retry once, return error if still failing
Timeout	Retry once; if it still times out, return an error
Content compliance interception	Return the compliance prompt directly without retrying
The user adds a resolution/super-resolution request midway	Return "This tool does not handle resolution-related requests"
The input is an image	Reject; this tool only supports video
"Restore to 4K" and similar cases that contain both restoration + resolution	This tool does not handle it (it includes a resolution request, which is outside the scope of its capabilities)

Deliver

Parse --json → downloaded_files[0].saved_path or media_urls[0]。

mv {file} {output_dir}/{YYYY-MM-DD}_{name}_quality-enhance.mp4

Output

• Format：MP4
• Naming：{YYYY-MM-DD}_{descriptive-name}_quality-enhance.mp4
• Location: Project → ./output/; One-time → $VISUAL/output/video-quality-enhance/

Baseline Task ID

t_mt1a3i5n7b4715ca50-43d6-4ace-beec-b2557b9700ed