editor-sfx-sparse
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted data from an external source and user requests to drive actions in an editor environment.
- Ingestion points: User-provided sound design requests and external metadata retrieved from the Freesound catalog via the
media_searchCatalogtool inSKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or specific warnings to the agent to ignore potential instructions embedded in catalog metadata.
- Capability inventory: The skill utilizes tools to search, import media, and modify timeline elements (
media.importFromCatalog,timeline.addMediaClip,element.setParams) within theSKILL.mdfile. - Sanitization: The skill recommends using specific
sfxRoletokens and search filters, which provides some implicit structure, but lacks explicit validation or escaping of the retrieved external content.
Audit Metadata