agent-memory
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Downloads the
agent-memoryrepository from GitHub and installs the@tencentdb-agent-memory/memory-tencentdbNode.js plugin. These resources originate from the skill author.\n- [COMMAND_EXECUTION]: Executes vendor-provided scripts including./start-all.shandmigrate-memory-to-tencentdb.mjsto manage services and data.\n- [PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by ingesting untrusted data from conversation history, codebases, and documentation.\n - Ingestion points: Raw conversation logs (L0), code symbols and relationships (CodeGraph), and project documentation (Wiki) are read from the file system.\n
- Boundary markers: The instructions do not define clear delimiters or "ignore embedded instructions" warnings for the ingested data.\n
- Capability inventory: Scripts associated with this skill utilize
Bash,Write,Read, andWebFetchtools.\n - Sanitization: There is no mention of sanitization, filtering, or validation of the external content before it is processed by the agent.
Audit Metadata