hiapi-gpt-image-2
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The installer (
scripts/install.mjs) utilizesgit cloneto retrieve source code from the official vendor repository on GitHub. - [EXTERNAL_DOWNLOADS]: The skill performs network requests to the HiAPI service (
api.hiapi.ai) and checks for updates by fetching a manifest from the vendor's GitHub repository (github.com/HiAPIAI/hiapi-skills). - [PROMPT_INJECTION]: The skill ingests untrusted data from users and remote sources, facilitating a surface for indirect prompt injection.
- Ingestion points: Prompt strings and external image URLs provided as arguments to
scripts/hiapi-gpt-image-2.mjs. - Boundary markers: None; prompts are interpolated directly into the JSON request body sent to the API.
- Capability inventory: The skill is capable of performing network requests and writing files to the local file system.
- Sanitization: The skill performs basic validation on the number and format of input URLs, but does not apply sanitization or delimiters to the text prompts.
Audit Metadata