Security audit

Daily Life Autopilot

Security checks for vulnerabilities and agentic risk

Overview

The skill is not overtly malicious, but it gives an AI broad automatic access to email, calendar, files, bills, and personal commitments without enough opt-in scoping.

Review this skill carefully before installing. It may be useful if you want a proactive assistant, but only use it with accounts and folders you are comfortable exposing, and prefer explicit opt-in settings for morning briefings, meeting briefs, billing tracking, weekly summaries, and file organization.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

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

Findings (7)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The skill advertises unrestricted natural-language activation with 'no commands to memorize,' which creates an overly broad control surface for a highly privileged agent that can access email, calendar, and files. In this context, ambiguous phrasing increases the chance of unintended invocation from ordinary conversation or injected content, causing the agent to read sensitive data or initiate workflows the user did not mean to trigger.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: Triggering on 'Any first message of the day' is dangerously broad because virtually any greeting or unrelated prompt can cause the agent to perform morning briefing behavior and access calendar, inbox, and follow-up data. For a life-management skill with broad personal-data access, this creates a substantial risk of unintended disclosure and over-collection without a clear, informed user action.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: Automatic activation on the first message of each day is insufficiently constrained and effectively grants the skill a standing right to inspect sensitive sources based on minimal user interaction. Because the module aggregates email, messages, tasks, and inferred commitments, accidental triggering could expose a large amount of personal information in contexts where the user only intended a simple conversation.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 85% confidence
Finding: This trigger does not define what qualifies as a billing email or whether forwarding alone constitutes consent to extract and track financial details. Broad phrasing around financial workflows can cause the skill to ingest more messages than the user intended.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 88% confidence
Finding: Automatically generating meeting briefs 30 minutes before any calendar event is too broad for a calendar that may contain personal, sensitive, or trivial events. This can cause unnecessary access to recent email threads and attendee context for events where such processing is not wanted, increasing privacy exposure and the chance of surfacing sensitive information at the wrong time.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The phrase 'Wrap up' is common everyday language and can easily appear in normal conversation unrelated to the end-of-day review workflow. In a skill that tracks commitments, meetings, and priorities, such a generic trigger can unintentionally launch a summary that reveals sensitive work or personal details.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: Running a weekly catch-up automatically every Monday morning without clear opt-in or scoping creates recurring autonomous access to inbox, billing, meetings, and file-activity data. Even if read-only, scheduled summaries can surface sensitive information unexpectedly and normalize background processing the user may not fully understand or intend.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.