elite-longterm-memory

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were detected. The skill operates as a utility for persistent context management using standard file operations and verified third-party service integrations.
  • [COMMAND_EXECUTION]: The skill includes a Node.js CLI tool (bin/elite-memory.js) used for initializing the memory structure and generating daily logs. Analysis confirms it only performs local file system operations within the workspace.
  • [EXTERNAL_DOWNLOADS]: Instructions recommend the installation of the mem0ai package and optional integration with the SuperMemory API. These are well-known services for AI agent development and do not involve suspicious or unverified sources.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (Category 8) inherent to its role as a memory system.
  • Ingestion points: SESSION-STATE.md, MEMORY.md, and daily logs in the memory/ directory.
  • Boundary markers: Absent; the system uses standard Markdown files for storage without explicit delimiters for untrusted content.
  • Capability inventory: The agent is instructed to read these files as primary context and write to them using standard file-system capabilities.
  • Sanitization: Absent; the system relies on the LLM's internal safety filters and logic when processing remembered information.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 04:30 PM