← Back to Skills Marketplace
517
Downloads
0
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install wavespeed-watermark-remover
Description
Remove watermarks, logos, captions, and text overlays from images and videos using WaveSpeed AI. Intelligently detects and removes watermarks while preservin...
Usage Guidance
This skill appears to do what it says (call a WaveSpeed API to remove watermarks), but the package metadata is incomplete: SKILL.md requires WAVESPEED_API_KEY and a 'wavespeed' client library while the registry entry lists no env vars or dependencies. Before installing, ask the publisher for: (1) a homepage or source repo to verify provenance; (2) explicit dependency/install instructions for the 'wavespeed' SDK; and (3) updated metadata declaring WAVESPEED_API_KEY as a required credential. Only provide an API key you trust to this third-party service and avoid hardcoding it — use a secret manager or per-project key. Test with non-sensitive media first, review the network endpoints the client uses, and consider legal/ethical implications of removing watermarks on copyrighted content.
Capability Analysis
Type: OpenClaw Skill
Name: wavespeed-watermark-remover
Version: 1.0.0
The skill is classified as suspicious due to the explicit warnings in the `SKILL.md` file's 'Security Constraints' section regarding 'No arbitrary URL loading' and 'Input validation' for media URLs. This indicates a significant vulnerability risk where the AI agent could be coerced into fetching content from untrusted or internal URLs (SSRF) or potentially local files (via `wavespeed.upload` if paths are user-controlled) if it does not rigorously validate user-provided inputs. While the skill's stated purpose is benign, it exposes a capability that, if misused by the agent, could lead to security compromises.
Capability Assessment
Purpose & Capability
The name and description match the SKILL.md instructions: it calls out image/video watermark removal via a WaveSpeed AI service. However the skill metadata declares no required environment variables or dependencies while the instructions explicitly require a WAVESPEED_API_KEY and import a 'wavespeed' client library — this mismatch suggests the registry metadata is incomplete.
Instruction Scope
The SKILL.md directs the agent to upload local media files and call a remote 'wavespeed' service using an API key. It does not instruct reading unrelated system files, but it does reference a local file upload flow and an environment variable (WAVESPEED_API_KEY) that are not declared in the registry metadata. The instructions also accept arbitrary URLs (with a security note advising against untrusted URLs) — the agent will need to access local files and external network endpoints, which is expected for this functionality but should be explicit in metadata.
Install Mechanism
There is no install spec (instruction-only), which is low risk. However the SKILL.md example imports a 'wavespeed' JS package and uses a Client API, but the skill does not declare that dependency or provide installation instructions. That omission is an inconsistency: the runtime will need that client library available but the skill gives no guidance on obtaining it.
Credentials
The SKILL.md requires a WAVESPEED_API_KEY for authentication (and shows how to export it), but the registry metadata lists no required environment variables or primary credential. Requesting one service API key is proportionate to the stated purpose, but the missing declaration in metadata is a meaningful discrepancy. There are no other unrelated credentials requested.
Persistence & Privilege
The skill does not request persistent presence (always is false), does not modify other skills or system config, and does not declare privileged config paths. Autonomous invocation is allowed (default) but is not combined with other alarming privileges.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install wavespeed-watermark-remover - After installation, invoke the skill by name or use
/wavespeed-watermark-remover - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
WaveSpeedAI Watermark Remover 1.0.0 – Initial Release
- Supports removing watermarks, logos, captions, and text overlays from images and videos using WaveSpeed AI.
- Preserves original texture and background with intelligent detection.
- Video watermark remover uses temporal-aware inpainting to prevent flickering, for videos up to 10 minutes.
- Provides simple API usage examples for both images and videos, plus advanced client options and error handling.
- Transparent pricing: $0.012 per image, $0.01 per video second (min $0.05).
- Lists security best practices, including proper API key handling and media input validation.
Metadata
Frequently Asked Questions
What is WaveSpeedAI Watermark Remover?
Remove watermarks, logos, captions, and text overlays from images and videos using WaveSpeed AI. Intelligently detects and removes watermarks while preservin... It is an AI Agent Skill for Claude Code / OpenClaw, with 517 downloads so far.
How do I install WaveSpeedAI Watermark Remover?
Run "/install wavespeed-watermark-remover" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is WaveSpeedAI Watermark Remover free?
Yes, WaveSpeedAI Watermark Remover is completely free (open-source). You can download, install and use it at no cost.
Which platforms does WaveSpeedAI Watermark Remover support?
WaveSpeedAI Watermark Remover is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created WaveSpeedAI Watermark Remover?
It is built and maintained by C (@chengzeyi); the current version is v1.0.0.
More Skills