Looki Memory

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Looki integration that can access very private wearable memory data, but its sensitive behavior is aligned with its stated purpose and is scoped to documented API use.

Install only if you trust Looki and the base URL you configure. Treat the API key like a password, prefer narrow date or topic requests, avoid saving the key if you do not need persistence, and remove ~/.config/looki/credentials.json if you stop using the skill.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill grants access to extremely sensitive lifelog, location, relationship, and realtime context data, but the description does not require an explicit per-use consent or privacy warning before retrieving that information. In an agent setting, this increases the risk of overbroad collection or disclosure of intimate personal data without the user fully understanding the sensitivity at the moment of access.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The setup instructs the agent to offer saving a long-lived API key to disk, but it does not clearly warn the user that the secret is persistent and highly sensitive. Users may consent without understanding that anyone or any process with access to that file could retrieve the key and access private memory data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal