mobile-storage-watermelondb
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents and implements a data synchronization protocol that ingests untrusted data from a remote API.
- Ingestion points: The
pullChangesfunction inSKILL.mdandexamples/sync.mdfetches data from an external URL. - Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the demonstrated sync logic to prevent the agent from following instructions embedded in the synced data.
- Capability inventory: The skill has the capability to write to the local file system (via SQLite) using
@writermethods and batch operations as described inexamples/core.mdandexamples/sync.md. - Sanitization: The skill mitigates risks by recommending the use of sanitizer functions with
@jsondecorators and providing theQ.sanitizeLikeString()utility for escaping user-provided search terms in queries.
Audit Metadata