remember

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a local state management system using 'memory.md'. It operates entirely within the filesystem of the current project and does not engage in network communication or external data transmission.
  • [PROMPT_INJECTION]: The skill's restoration process ingests data from local files which constitutes a potential surface for indirect prompt injection if project files are compromised.
  • Ingestion points: The agent reads 'memory.md', 'CLAUDE.md', and other project context files during the 'restore' command.
  • Boundary markers: Absent; the skill reads file content directly into the context without specific delimiters or isolation instructions.
  • Capability inventory: File system access (read/write) consistent with a standard developer agent role.
  • Sanitization: No explicit data sanitization is performed on the ingested file content; however, the skill includes a mandatory human-in-the-loop verification step where the agent must summarize the restored state and wait for user confirmation before proceeding.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 08:42 PM
Security Audit — agent-trust-hub — remember