Generate product photos for ecommerce

v1.0.1

Generate professional AI product photos using ProductAI.photo service. Use when users need to create, enhance, or transform product images for e-commerce, ma...

0· 383·1 current·1 all-time
byMarko Balažic@shapes
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (generate product photos) matches the included files and runtime behavior. The package contains a client, setup helper, and CLI scripts that call a ProductAI API endpoint (api.productai.photo). No unrelated cloud credentials, system services, or external tools are required.
Instruction Scope
SKILL.md and scripts instruct the agent/user to run an interactive setup that stores a ProductAI API key in ~/.openclaw/workspace/productai/config.json and then call API endpoints to generate/upscale images. That is within scope. Note: the repository includes programmatic setup examples that would let an agent write the API key on behalf of a user (expected for this integration, but users should be careful when sharing keys in chat).
Install Mechanism
This is instruction-only with no automated install spec. Python scripts are provided and require standard Python deps (requests, pillow). No downloads from unknown URLs or archive extraction are present.
Credentials
The integration only needs the ProductAI API key and local config file; no unrelated environment variables or broad credential sets are requested. The API key is stored in a user-scoped config file (permission 600) as documented. No other secrets are requested.
Persistence & Privilege
Skill does not request always:true, does not modify global agent settings beyond writing its own config.json, and requires user action to set up the API key. Autonomous invocation is allowed by default (platform normal), but the skill itself doesn't escalate privileges.
Assessment
This skill appears to do what it claims (a ProductAI.photo client and CLI). Before installing or handing over an API key, consider the following: - Don't paste a production API key into an untrusted chat or agent; prefer a test key or token with limited quota while validating the skill. Rotate the key if it is exposed. - The setup saves your API key to ~/.openclaw/workspace/productai/config.json with 0o600 permissions (script enforces this). That is standard for CLI tools but still means any local process with your user privileges can read it. - The code performs basic SSRF protection by blocking HTTP and obvious private hostnames, but hostname-only checks can be bypassed if a hostname resolves to a private IP. If you will supply untrusted input for image URLs, review or harden URL-to-IP resolution checks (resolve DNS and verify IP ranges) before production use. - Generated image URLs returned by the API are downloaded without additional URL validation; if you have concerns, inspect returned URLs before download or add validation for domains you trust. - Minor inconsistencies exist (e.g., setup.py default model string differs from other docs: 'nano-banana-2' vs 'nanobanana', and package.json lists Python deps in an npm manifest). These look like documentation/config mismatches rather than malicious behavior, but you may want to correct them. Recommended steps before trusting this skill: review the productai_client.py and setup.py locally, test with a low-privilege/test API key, monitor your ProductAI dashboard for unexpected usage, and consider applying additional SSRF protections if you will process untrusted image URLs.

Like a lobster shell, security has layers — review code before you run it.

latestvk97ax889s3bkcm8rjxx09may0181rbpe

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments