Skills video-inpainting

📦

video-inpainting

Name: video-inpainting
Author: runcomfy-com

Content revision r1 Medium Risk ⚙️ External commands🌐 Network access📁 Filesystem access

Edit Video Regions with RunComfy

Video region edits can be hard to route across models and frame methods. This skill selects a RunComfy path and gives clear prompts for object removal, cleanup, and replacement.

Supports: Claude Codex Code(CC)

📊 71 Adequate

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "video-inpainting" from https://skillstore.io/skills/runcomfy-com-video-inpainting.md and its manifest at https://skillstore.io/api/skills/runcomfy-com-video-inpainting/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/runcomfy-com-video-inpainting.md Signed manifest GET /api/skills/runcomfy-com-video-inpainting/manifest Signed lockfile GET /api/skills/runcomfy-com-video-inpainting/lockfile

Test it

Using "video-inpainting". Remove a small logo from the bottom-right corner of an owned clip.

Expected outcome:

Recommended route: Wan 2-7 Edit-Video. The prompt targets the bottom-right logo, preserves all other content, and saves results to the output folder.

Using "video-inpainting". Change a jacket color while keeping the person recognizable.

Expected outcome:

Recommended route: Lucy Edit Restyle. The guidance focuses on identity stability, one region change, and checking frames for drift.

Using "video-inpainting". Edit a short low-frame-rate sequence frame by frame.

Expected outcome:

Recommended route: Seedream 4-0 Edit-Sequential. The plan treats frames as still images, then advises re-encoding only after review.

Security Audit

Medium Risk

v4 • 7/5/2026 Open versioned report

No prompt injection attempt was found in SKILL.md. The static external command hits are Markdown backticks and bounded RunComfy CLI documentation, while the network hits are visible links or placeholder URLs. The main remaining issue is semantic misuse risk from explicit watermark and logo removal guidance.

Files scanned

176

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (1)

Medium

SKILL.md:15-18 SKILL.md:113-115

Watermark Removal Misuse Risk

The skill explicitly supports watermark and logo removal workflows. This can enable unauthorized removal of ownership, licensing, or attribution marks from third-party media.

The skill directly lists watermark removal triggers and a common watermark/logo removal pattern. Legitimate cleanup is possible, so this is a misuse risk rather than clear malicious intent.

Risk Factors

⚙️ External commands (31)

🌐 Network access (21)

📁 Filesystem access (2)

SKILL.md:160 SKILL.md:160

Audited by: codex View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/runcomfy-com-video-inpainting/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/runcomfy-com-video-inpainting/security.svg)](https://skillstore.io/skills/runcomfy-com-video-inpainting?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/runcomfy-com-video-inpainting?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/runcomfy-com-video-inpainting/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/runcomfy-com-video-inpainting.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

runcomfy-com. (2026). video-inpainting security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/runcomfy-com-video-inpainting/audits/4

BibTeX citation

@techreport{runcomfy-com-runcomfy-com-video-inpainting-2026,
  author = {runcomfy-com},
  title = {video-inpainting security audit report (audit version 4)},
  institution = {Skillstore},
  year = {2026},
  number = {4},
  url = {https://skillstore.io/skills/runcomfy-com-video-inpainting/audits/4},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "video-inpainting security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
  - name: "runcomfy-com"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/runcomfy-com-video-inpainting/audits/4"
identifiers:
  - type: other
    value: "skillstore:runcomfy-com-video-inpainting:audit:4"
    description: "Skillstore immutable audit report identifier"

Compare variants

3 installable variants

Each author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.

Why this variant is first

Highest Skillstore Score

Author / source Skillstore Score Evidence Confidence Skillstore usage Updated

agentspace-so Recommended

agentspace-so-video-inpainting

Skillstore Score 78

Evidence Confidence High

Skillstore usage 10

Updated

2026-08-21

runcomfy-com Current

runcomfy-com-video-inpainting

Skillstore Score 71

Evidence Confidence High

Skillstore usage 10

Updated

2026-08-21

doany-ai

doany-ai-video-inpainting

Skillstore Score 70

Evidence Confidence High

Skillstore usage 8

Updated

2026-08-21

Skillstore Score

Why this score Evidence Confidence: High

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Clean Up Owned Footage

Remove a distracting object, cable, or blemish from video that the user owns or has permission to edit.

Prepare Product Clips

Replace or refine a visible region while keeping the rest of a product clip stable across frames.

Prototype Visual Effects

Compare prompt-driven edits with frame-stack or ComfyUI routes before committing to a final workflow.

Try These Prompts

Remove One Object

Use video-inpainting to remove the small object in my clip. Keep all other content unchanged and recommend the simplest RunComfy route.

Choose the Model Route

Review my edit goal and choose between Wan 2-7, Lucy Edit Restyle, and Seedream Edit-Sequential. Explain the tradeoffs briefly.

Write a Preservation Prompt

Draft a RunComfy video inpainting prompt for this region edit. Lead with preservation, name the target region, and keep it to one change.

Plan a Mask-Precise Workflow

Plan the safest workflow for a clip that needs exact mask propagation. Include when to leave the CLI and use a RunComfy ComfyUI workflow.

Best Practices

Use media you own or have permission to edit.
Describe one region and one change per run.
Use ComfyUI workflows when masks or temporal consistency must be precise.

Avoid

Do not remove ownership marks from third-party media without permission.
Do not combine many unrelated edits in one prompt.
Do not use frame-stack editing for long clips that need motion coherence.

Frequently Asked Questions

Does this skill perform the video edit itself?

It guides and invokes RunComfy CLI routes when the CLI is available. RunComfy performs the cloud processing.

Which model is the default?

Wan 2-7 Edit-Video is the default for prompt-driven region edits across video frames.

Can it use exact masks?

The CLI routes are prompt-driven. Exact mask propagation requires a RunComfy ComfyUI workflow.

Where are results saved?

The RunComfy CLI downloads completed outputs to the output directory chosen for the run.

Does it send videos to RunComfy?

Yes. The workflow sends the user-approved video URL and prompt to RunComfy for processing.

Can it remove watermarks?

Only use that workflow for media you own or have clear permission to edit.

Developer Details

Author

runcomfy-com

License

MIT

Skillstore revision

Version notice

The author did not declare a version.

Repository

https://github.com/runcomfy-com/skills/tree/main/video-inpainting/

Ref

62e2a730c5cd74eab4c7164309d810de660fcea3

Maintenance freshness

7/18/2026

Usage

4 downloads · 0 views

File structure

📄 SKILL.md