Evolink Nano Banana 2 1

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This is a coherent image-generation skill, but users should notice that it uses an Evolink API key, may upload prompts/images to Evolink, and recommends an unpinned npm MCP setup command.

This skill appears purpose-aligned for Evolink image generation. Before installing, confirm the publisher/package identity, use a dedicated Evolink API key, consider pinning the MCP npm package instead of using @latest, and avoid uploading sensitive images or prompts unless you are comfortable sending them to Evolink.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI03: Identity and Privilege Abuse
Low
What this means

Anyone installing this skill must provide or expose an Evolink API key, which can be used to consume credits or access Evolink-hosted files within that account's permissions.

Why it was flagged

The skill needs a service API key to authenticate requests to Evolink. This is expected for the integration, but it gives the skill delegated access to the user's Evolink account.

Skill content
`EVOLINK_API_KEY` authenticates all requests. Injected by OpenClaw automatically. Treat as confidential.
Recommendation

Use a dedicated Evolink API key, monitor usage and billing, and revoke the key if you stop using the skill.

#
ASI07: Insecure Inter-Agent Communication
Medium
What this means

Private prompts or sensitive images submitted for editing/generation will be sent to a third-party service and may be accessible through temporary hosted links.

Why it was flagged

The skill clearly discloses that user prompts and image inputs are transmitted to Evolink and that generated or uploaded content is available through temporary URLs.

Skill content
Prompts and images are sent to `api.evolink.ai`. Uploaded files expire in **72h**, result URLs in **24h**.
Recommendation

Do not submit confidential, regulated, or highly personal images unless Evolink's privacy, retention, and access controls meet your needs.

#
ASI04: Agentic Supply Chain Vulnerabilities
Low
What this means

If the npm package changes or is compromised later, a user following the setup command could run different code than what was reviewed here.

Why it was flagged

The setup documentation recommends running an external npm MCP package with the moving `@latest` tag. This is purpose-aligned setup guidance, not automatic execution, but the exact code version is not pinned in the artifact.

Skill content
`mcporter call --stdio "npx -y @evolinkai/evolink-media@latest" list_models`
Recommendation

Verify the npm/GitHub package publisher and consider pinning a specific trusted version instead of using `@latest`.

#
ASI04: Agentic Supply Chain Vulnerabilities
Info
What this means

The skill identity metadata is not perfectly consistent, so users may want to confirm they are installing the intended Evolink skill.

Why it was flagged

The packaged metadata differs from the registry-provided owner and slug shown in the evaluation context. This is a provenance inconsistency, though it does not by itself show unsafe behavior.

Skill content
"ownerId": "kn74p4xy6sja0199cea53anecs81kqjs", "slug": "evolink-nano-banana-2"
Recommendation

Verify the publisher, homepage, and package identity before providing an API key.