tencentdb-agent-memory
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill provides installation commands for the OpenClaw plugin (
openclaw plugins install) and Docker build/run commands. These are standard operational instructions for the claimed purpose. - [REMOTE_CODE_EXECUTION]: The documentation mentions running a shell script
bash scripts/openclaw-after-tool-call-messages.patch.sh. This script is part of the skill's own local repository for patching the host environment, which is a common practice for plugin integrations. - [EXTERNAL_DOWNLOADS]: The skill fetches the official TencentDB Agent Memory library from NPM (
@tencentdb-agent-memory/memory-tencentdb). This is a legitimate package for the skill's described functionality. - [DATA_EXPOSURE]: The skill correctly advises users to manage sensitive information like
LLM_API_KEYandPOSTGRES_PASSWORDusing environment variables and.envfiles, which aligns with security best practices. - [SAFE]: The memory extraction and symbolic compression (using Mermaid graphs and local SQLite storage) are handled entirely on the local system, minimizing external data exposure risks.
Audit Metadata