AgentDB Memory Patterns
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation suggests using
npx agentdb@latestto initialize databases and run Model Context Protocol (MCP) servers. These tools are associated with the author's ecosystem and are standard for the advertised functionality. - [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for storing and retrieving external interaction data, which constitutes a potential attack surface for indirect prompt injection.
- Ingestion points: Untrusted data is ingested into the persistent store via methods such as
insertPattern,storeMemory, andstoreFactdescribed inSKILL.md. - Boundary markers: The code examples do not explicitly demonstrate the use of delimiters or 'ignore embedded instructions' warnings for stored content.
- Capability inventory: The skill focuses on data persistence and retrieval; no arbitrary command execution or network exfiltration capabilities are exposed in the provided patterns.
- Sanitization: There is no explicit evidence of sanitization or validation of the content before it is stored or retrieved for agent context.
Audit Metadata