← Back to Skills Marketplace
140
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install imglink
Description
Generate images by customizing a URL. Drop the URL into websites, presentations, PDFs, or anywhere that loads images.
README (SKILL.md)
Use imglink.ai to generate images via URL. The API is a single GET endpoint that accepts a prompt parameter. It returns the image.
GET https://imglink.ai/images?prompt=cat
Parameters
- prompt (required): Text description of the image to generate. URL-encode spaces as +.
- width (optional, default 800): Image width in pixels.
- height (optional, default 600): Image height in pixels.
- version (optional, default 1): Integer seed for deterministic output. Same prompt + version always returns the same image.
- model (optional, default "nano-banana-2"): AI model to use for generation.
- key (optional, default "anonymous"): API key. Use "anonymous" for testing (rate-limited).
Rules
- Every unique URL returns a real generated image. Use these URLs directly in \x3Cimg> tags, Markdown, CSS, or anywhere an image URL is accepted.
- Images are generated on first request, then cached permanently. Same prompt + version = same image.
- URL-encode the prompt value. Use + for spaces.
- Do not invent parameters that are not listed above.
Examples
\x3Cimg src="https://imglink.ai/images?prompt=frog&key={key}" />
\x3Cimg src="https://imglink.ai/images?prompt=dog&width=1024&height=768&version=1&model=nano-banana-2&key={key}" />
Getting a key
A key can be obtained for free by logging in at https://imglink.ai - without a key, responses are heavily rate-limited.
Notes
First response will take longer as the image is generated. Future fetches load from cache.
Usage Guidance
This skill is coherent and only documents how to build image-generation URLs for the external domain imglink.ai. Before using it, verify the external service (imglink.ai) and its privacy/terms because: generated images are said to be cached permanently (so avoid sending sensitive or private content in prompts), the optional 'anonymous' key is rate-limited (get a real key if you need higher throughput), and the skill has no published homepage or source—if you plan to use it in production, check the service's reputation, rate limits, pricing, and legal/ownership issues for generated images. Do not include secrets or PII in prompts you send to the service.
Capability Analysis
Type: OpenClaw Skill
Name: imglink
Version: 1.0.2
The skill is a documentation-only bundle providing instructions for an AI agent to interact with the imglink.ai image generation API. It contains no executable code or scripts, and the instructions in SKILL.md are strictly limited to constructing valid API URLs for image generation. No indicators of malicious intent, data exfiltration, or harmful prompt injection were found.
Capability Assessment
Purpose & Capability
Name/description (generate images via URL) match the SKILL.md: it documents a single public GET endpoint and parameters needed to produce image URLs. There are no unrelated credentials, binaries, or config requirements.
Instruction Scope
SKILL.md only instructs constructing and using a public HTTPS GET URL (prompt, width, height, model, version, key). It does not tell the agent to read files, credentials, system state, or send data to unexpected endpoints. It does warn that images are cached permanently — an important behavioral note but within scope.
Install Mechanism
No install spec and no code files (instruction-only). Nothing will be downloaded or written to disk by the skill itself.
Credentials
No required environment variables or credentials declared. The SKILL.md references an optional API key obtained from the service (expected and proportional).
Persistence & Privilege
always is false and the skill does not request persistent system-level privileges or modify other skills. Autonomous invocation is allowed (platform default) but not combined with other red flags.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install imglink - After installation, invoke the skill by name or use
/imglink - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- Updated SKILL.md example section to use code blocks for image URL usage.
- No changes to API functionality or parameters.
v1.0.1
- Updated documentation for clarity and conciseness.
- Simplified the API call example to show a direct GET request using only the required prompt parameter.
- Removed redundant example using plain URLs and focused examples on <img> tags.
- No changes to the API functionality or parameters.
v1.0.0
Initial release of imglink skill:
- Generate images by customizing a URL with prompt and optional parameters.
- Supports direct image embedding in websites, presentations, and documents.
- Deterministic outputs: same prompt and version produce the same image.
- Built-in parameters for prompt, width, height, version, model, and API key.
- Example URLs and usage instructions included.
Metadata
Frequently Asked Questions
What is ImgLink?
Generate images by customizing a URL. Drop the URL into websites, presentations, PDFs, or anywhere that loads images. It is an AI Agent Skill for Claude Code / OpenClaw, with 140 downloads so far.
How do I install ImgLink?
Run "/install imglink" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is ImgLink free?
Yes, ImgLink is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does ImgLink support?
ImgLink is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created ImgLink?
It is built and maintained by jverlee (@jverlee); the current version is v1.0.2.
More Skills