personize-memory

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provided recipes and templates utilize standard third-party libraries for data integration, including @personize/sdk, @hubspot/api-client, jsforce, pg, and mysql2.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process untrusted data from external sources for AI extraction.\n
  • Ingestion points: Untrusted content enters the agent context through the memorize and memorizeBatch methods in SKILL.md and recipes/data-sync.ts.\n
  • Boundary markers: The documentation does not specify the use of delimiters or 'ignore' instructions for the content being processed.\n
  • Capability inventory: The skill utilizes network operations for API interaction and performs data storage across its integration templates.\n
  • Sanitization: External content is processed for semantic extraction without specific sanitization filters described in the instructions.\n- [SAFE]: No malicious patterns, unauthorized data exfiltration, or obfuscation were detected. The skill follows standard security practices for managing sensitive information via environment variables.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 04:20 PM