Security audit

augmentiq

Security checks for vulnerabilities and agentic risk

Overview

AugmentiQ is a clearly disclosed local memory skill, but users should understand that it persistently stores selected conversation context in an Obsidian vault.

Install this only if you want an agent to keep long-term local memory in your Obsidian vault. Prefer read-only mode or narrow vault access if you do not want note mutation, avoid sharing secrets, and periodically review or clear stored memories and profile entries.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The skill instructs agents to use the memory system in every main-session conversation, creating an always-on collection pattern without a narrowly scoped activation rule or explicit per-conversation consent. That increases the chance of unnecessary retention of personal or sensitive context and makes accidental over-collection more likely.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill markets persistent memory and evolving profiles but does not prominently warn that user conversations and inferred attributes may be written to local storage by default. Even though storage is local, silent persistence of personal data is a privacy risk because users may not realize their interactions are being retained across sessions.

Ssd 3

Medium

Confidence: 93% confidence
Finding: The skill explicitly directs persistent collection and summarization of user information across all main sessions, normalizing broad profiling behavior as a default. In a memory skill, this materially raises privacy risk because it encourages continuous accumulation of user data beyond what is strictly necessary for the immediate task.

Ssd 3

Medium

Confidence: 95% confidence
Finding: The session summary guidance tells the agent to retain broad conversation details, decisions, action items, mood, and anything the user asked to remember. This creates a high likelihood of storing sensitive behavioral and emotional data that may not be necessary, increasing privacy exposure and the blast radius of any local compromise or misuse.

Ssd 3

Medium

Confidence: 96% confidence
Finding: The first-run onboarding encourages soliciting personal details and storing each item with maximum importance, which promotes aggressive and durable retention from the start. That is dangerous because it can overclassify trivial or sensitive personal information as permanent memory, making later correction, minimization, or deletion less likely.

Static analysis

No suspicious patterns detected.