spacetimedb-clients
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents how to build applications that synchronize and process data from remote database tables, creating a potential surface for indirect prompt injection if the database content originates from untrusted sources.
- Ingestion points: Data enters the agent's processing context via subscription callbacks (
onInsert,onUpdate) described inreferences/clients-api.mdandreferences/clients-subscriptions.md. - Boundary markers: The documentation does not provide instructions for implementing boundary markers or delimiters to separate untrusted data from instructions.
- Capability inventory: The skill documentation (specifically
references/clients-codegen.md) outlines capabilities for invoking reducers and procedures, including functions that can perform external HTTP requests. - Sanitization: There are no mentions of data validation or sanitization protocols for the ingested database rows before they are used to influence application logic or reducer calls.
Audit Metadata