happyhorse
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references official installation instructions and service-specific extensions from GitHub repositories associated with the
inference-shandbelt-shorganizations. These references are essential for the operation of the CLI tool required by the skill. - [INDIRECT_PROMPT_INJECTION]: The skill includes an attack surface for indirect prompt injection as it processes external, potentially untrusted data like text prompts and media URLs.
- Ingestion points: Data enters the system via
prompt,reference_images, andvideoparameters within thebelt app runcommands documented inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the documentation.
- Capability inventory: The skill uses the
Bash(belt *)tool, which performs network operations to interact with remote AI inference services. - Sanitization: There is no evidence of input validation or sanitization for the strings and URLs provided to the tool.
Audit Metadata