tanstack-ai-memory-hindsight
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the
@tanstack/ai-memoryand@vectorize-io/hindsight-clientNode.js packages as dependencies for the hosted memory adapter. - [PROMPT_INJECTION]: The skill provides tools for long-term memory (
hindsight_recall,hindsight_retain), which represents a potential surface for indirect prompt injection if malicious data is stored and later recalled. - Ingestion points: Data is ingested and stored in the memory bank via the
hindsight_retaintool during conversations. - Boundary markers: The provided documentation does not specify the use of delimiters or boundary markers for the data stored in memory.
- Capability inventory: The skill enables the agent to autonomously manage memory through
retain,recall, andreflecttools. - Sanitization: No explicit sanitization or filtering mechanisms for stored or retrieved content are described in the skill instructions.
Audit Metadata