memory-graph

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires running an installation script (assets/install.sh) that unzips a bundled binary (memory-graph) and installs it to the user's local directory (~/.local/bin). This is a standard installation procedure for this tool.
  • [DATA_EXPOSURE]: The skill uses /tmp/memory-apply.yaml to store temporary data for user review. While typical for temporary storage, users should be aware that /tmp is a shared system directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill parses conversation history to提炼 (extract) conclusions and rules. It mitigates potential injection risks by implementing a mandatory human-in-the-loop step, requiring the user to review and confirm the extracted data before it is written to the global memory store.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 03:12 PM
Security Audit — agent-trust-hub — memory-graph