Pregnancy (Tracker, Journal, Triage, Visit Prep)

• Outcome: Pass
• Latest audit: Sep 11, 2026 · 4:09 AM UTC
• Version: 1.0.0

Overview

This is a local, instruction-only pregnancy tracker that stores sensitive health notes only with user confirmation and does not request credentials or network access.

Before installing, understand that this skill may create plaintext local pregnancy-tracking files in ~/pregnancy/. Use it only on a trusted device or protected storage location, review planned writes before approving them, and rely on clinicians or emergency services for urgent symptoms or medical decisions.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• System Prompt Leakage: Direct Leakage, Indirect Extraction, Tool-Based Exfiltration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (7)

Direct Prompt Extraction

High

• Category: System Prompt Leakage

Content

Only add a new field when current completion stays stable for at least 5 days.
If completion drops, remove lowest-value fields first.

## Output Rule

Each output should include:
- what changed since last check-in

• Confidence: 85% confidence
• Finding: Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The 'When to Use' section describes the skill as applicable whenever a user wants tracking for symptoms, routines, medications, appointments, questions, or warning signs, which is a wide set of common health-related intents. It does not provide specific trigger phrases, scope constraints, or negative examples to clarify when this skill should not activate versus other general health skills.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The template instructs creation of multiple local files containing highly sensitive health information, including pregnancy status, symptoms, medications, vitals, fetal movement, appointments, and alert events, but provides no privacy, access-control, retention, or device-security guidance. This can lead users or downstream agents to persist regulated-like health data in plaintext on shared, synced, or insufficiently protected systems, increasing risk of unauthorized disclosure.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The activation scope is defined very broadly around pregnancy, prenatal care, symptoms, appointments, or baby movement, which can cause the skill to trigger in ordinary or tangential conversation. In a health-related skill, unintended activation is risky because it may surface medical-style guidance or begin collecting sensitive context without sufficiently clear user intent.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill states that it will store sensitive pregnancy-related data, including pregnancy week, warning events, and medical visit questions, but does not clearly present this as a retention notice to the user or obtain informed consent for storage itself. Because pregnancy and symptom history are highly sensitive health data, undisclosed retention increases privacy, trust, and misuse risks if data is stored or reused beyond the user's expectations.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: This pregnancy skill includes symptom/event tracking and 'immediate red or amber triage checks' but does not warn that the guidance is not a substitute for professional medical advice or emergency care. In a prenatal context, users may rely on the tool for time-sensitive decisions about bleeding, fluid leak, contractions, or fetal movement changes, which can delay urgent evaluation.

Scope Creep

Low

• Category: Excessive Agency

Content

- Mixing reassurance with warning signs -> delayed urgent care.
- Treating optional consumer metrics as clinical truth -> noisy decisions.
- Summaries with raw dumps only -> poor usability during appointments.
- Giving treatment advice beyond scope -> safety and trust risk.

## External Endpoints

• Confidence: 70% confidence
• Finding: Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Static analysis

No suspicious patterns detected.