Security audit

Memory LanceDB

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed long-term memory plugin that stores and searches conversation memories, and the sensitive behavior I found matches that purpose.

Install only if you want persistent long-term memory. Review autoCapture, autoRecall, dbPath, storageOptions, embedding provider, API key, and baseUrl settings because remembered text and recall queries may be stored locally or remotely and sent to the configured embedding service.

SkillSpector

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

Static analysis

Detected: suspicious.dangerous_exec, suspicious.dynamic_code_execution, suspicious.env_credential_access

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: dist/.setup/dist-A24Sj2xJ.mjs:84
Evidence: return __require("child_process").execSync("ldd --version", { encoding: "utf8" }).includes("musl");

Dynamic code execution detected.

Critical

Code: suspicious.dynamic_code_execution
Location: dist/.setup/dist-A24Sj2xJ.mjs:7696
Evidence: transformers = await eval("import(\"@huggingface/transformers\")");

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: dist/.setup/dist-A24Sj2xJ.mjs:90
Evidence: if (process.env.NAPI_RS_NATIVE_LIBRARY_PATH) try {