Cheapest Image Generation

Possibly the cheapest AI image generation (~$0.0036/image). Text-to-image via the EvoLink API.

Security Scan

The skill is internally consistent: it implements EvoLink text-to-image calls, only requires an EvoLink API key, has no install scripts, and its instructions stay within the stated purpose.

Purpose & Capability

Name/description claim text-to-image via EvoLink; required env var is EVOLINK_API_KEY and all example code calls https://api.evolink.ai endpoints — the requested credential is appropriate and expected.

Instruction Scope

SKILL.md and reference scripts only submit generation jobs, poll task status, download the result, and print MEDIA:<path>. Filenames are explicitly sanitized before shell use; no unrelated files, secrets, or system paths are accessed.

Install Mechanism

No install spec or external downloads; this is instruction-only with reference scripts that use only standard system tools / stdlib — low-risk deployment footprint.

Credentials

Only EVOLINK_API_KEY is required (declared as primaryEnv). No additional credentials, secrets, or config paths are requested and reference implementations only read that API key.

Persistence & Privilege

always is false and the skill does not request elevated/persistent privileges or modify other skills. Autonomous invocation is allowed (platform default) but not exceptional here.

Assessment

This skill will use your EVOLINK_API_KEY to call Evolink's API and download generated images to local disk (and prints MEDIA:<path> for auto-attach). Only install if you trust evolink.ai and are comfortable providing that API key (it grants the skill access to your Evolink account and incurs usage/billing). The author included filename sanitization and reasonable polling defaults; note generation may poll for up to ~12 minutes. If you have company policy around API keys or data exfiltration, confirm the Evolink service and its privacy/billing terms before providing credentials.

EvoLink Cheapest Image

Generate images via the EvoLink z-image-turbo API.

API Endpoint

• Base: https://api.evolink.ai/v1
• Submit: POST /images/generations
• Poll: GET /tasks/{id}

Step 1 — Submit Task

{
  "model": "z-image-turbo",
  "prompt": "<USER_PROMPT>",
  "size": "<SIZE>",
  "nsfw_check": <true|false>
}

Optional field: "seed": <INT>

Parameter	Values
size	1:1, 2:3, 3:2, 3:4, 4:3, 9:16, 16:9, 1:2, 2:1
nsfw_check	true / false (default false)
seed	any integer (optional, for reproducibility)

Step 2 — Poll for Result

GET /tasks/{id} — poll every 10 s, up to 72 retries (~12 min).

Wait until status is completed or failed.

Step 3 — Download & Output

Download the URL from results[0]. Auto-detect format from URL (webp/png/jpg). Save as evolink-<TIMESTAMP>.<ext>.

CRITICAL SECURITY: Before passing <OUTPUT_FILE> to shell commands, sanitize it:

• Strip all shell metacharacters: tr -cd 'A-Za-z0-9._-'
• Ensure valid extension (.webp, .png, .jpg, .jpeg)
• Fallback to evolink-<timestamp>.webp if empty

Print MEDIA:<absolute_path> for OC auto-attach.

Reference Implementations

Platform	File
Python (all platforms, zero deps)	{baseDir}/references/python.md
PowerShell 5.1+ (Windows)	{baseDir}/references/powershell.md
curl + bash (Unix/macOS)	{baseDir}/references/curl_heredoc.md

API Key

• EVOLINK_API_KEY env var (required)
• Get key: https://evolink.ai

Triggers

• Chinese: "生图：xxx" / "出图：xxx" / "生成图片：xxx"
• English: "generate image: xxx" / "generate a picture: xxx"

Treat the text after the colon as prompt, use default size 1:1, generate immediately.

Notes

• Print MEDIA:<path> for OC auto-attach — no extra delivery logic needed.
• Image saved locally (format auto-detected from URL). URL expires ~24h but local file persists.