apimodels-image
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes local Node.js scripts (
scripts/generate.mjsandscripts/build_catalog.mjs) to manage interactions with the service API. These scripts are self-contained, use built-in Node.js modules, and do not introduce external dependencies. - [CREDENTIALS_UNSAFE]: The skill follows strict security protocols for API key handling. It encourages the use of environment variables or a local configuration file (
~/.apimodels/credentials) with restricted access permissions (0600). The documentation explicitly instructs the agent and the user to avoid pasting secrets into the chat interface to prevent them from being logged in conversation histories. - [EXTERNAL_DOWNLOADS]: The tool downloads generated image files from the service provider's API to the local system when requested by the user. It also implements an RFC-compliant loopback authorization flow for secure authentication via a browser.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-supplied prompts for image generation, it correctly treats this data as input for the external API and does not execute the resulting data. It also includes logic to handle and report content moderation failures from the upstream provider.
Audit Metadata