Security audit

amem

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed memory plugin that stores and manages agent memories using local Qdrant and configured LLM providers, with no evidence of hidden exfiltration or deceptive behavior.

Before installing, users should understand that enabling conversation access allows this plugin to send memory-worthy conversation snippets to the configured LLM endpoint and store derived memories in local Qdrant. Keep API keys, Qdrant, model cache, and review-output directories under your control, and leave automatic conversation write-back disabled unless you want long-term memory from chats.

SkillSpector

SkillSpector was not run because this plugin release contains no bundled skills.

Static analysis

Detected: suspicious.env_credential_access

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: dist/index.js:55
Evidence: _dataDir = process.env.AMEM_DATA_DIR || path.join(os.homedir(), ".amem");