product-changelog
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references installation instructions for the belt CLI hosted on GitHub (inference-sh/skills). These references are for documentation purposes to help the user set up the required environment.
- [COMMAND_EXECUTION]: The skill provides examples of executing commands via the belt CLI to run specific AI applications (e.g., falai/flux-dev-lora, bytedance/seededit-3-0-i2i). These commands are intended to assist the user in generating visual assets for their changelogs.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text for release notes.
- Ingestion points: Product update descriptions and feature details provided by the user in SKILL.md.
- Boundary markers: The skill provides formatting templates but does not specify explicit delimiters for user content.
- Capability inventory: The skill utilizes belt app run to invoke remote tools based on the generated content.
- Sanitization: No explicit sanitization or filtering is described for the input data.
- [SAFE]: The skill operates within its stated purpose of document generation and visual asset creation. All external tools and sources referenced are legitimate services within the AI development ecosystem.
Audit Metadata