Nutrition Advisor

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 9:47 AM UTC
• Version: 1.0.1

Overview

This nutrition skill is purpose-aligned and disclosed, but users should be careful before saving health and diet logs locally.

Installers should treat saved nutrition profiles and daily logs as sensitive personal data. Use session-only help unless you need long-term tracking, and if you enable persistence, understand that the files may be readable by anything with access to your home directory or backups.

A.I.G

Vulnerability Patterns

• Insecure Skill Coding Practices: Finds exploitable flaws such as hardcoded secrets or command injection
• 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

Findings (1)

T09 · Insecure Skill Coding Practices

Warning

• Location: SKILL.md:143

Finding

Sensitive Health and Nutrition Data Stored in Predictable Plaintext Files

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, lines 143-225
Vulnerability Type: Plaintext storage of sensitive user data
Risk Level: Medium

The Skill instructs the Agent to persist body measurements, menstrual-cycle information, allergies, dietary preferences, nutrition targets, and meal history in unencrypted JSON files at predictable locations.

text

## User Profile Storage

When the user wants persistence, store profile data locally:

```text
~/nutrition-data/profile.json

Use this structure:

json

{
  "sex": "female",
  "age": 29,
  "height_cm": 168,
  "weight_kg": 62,
  "body_fat_percent": 25,
  "activity_level": "moderate",
  "goal": "fat_loss",
  "cycle": {
    "average_length_days": 28,
    "last_period_start": "2026-06-01"
  },
  "allergies": [],
  "intolerances": [],
  "disliked_foods": [],
  "dietary_preferences": []
}

Before writing or updating local files, explain what will be stored and ask for confirmation.

Daily Food Log

Store daily logs locally by date:

text

~/nutrition-data/daily-logs/YYYY-MM-DD.json

Use this structure:

json

{
  "date": "2026-06-28",
  "target": {
    "calories": 1650,
    "protein": 120,
    "carbs": 150,
    "fat": 50,
    "fiber": 30
  },
  "meals": [
    {
      "id": "m1",
      "type": "breakfast",
      "time": "08:15",
      "items": [
        {
          "name": "oats, dry",
          "amount_g": 40,
          "calories": 156,
          "protein": 6.8,
          "carbs": 26.4,
          "fat": 2.8,
          "fiber": 4.1
        }
      ],
      "total": {
        "calories": 156,
        "protein": 6.8,
        "carbs": 26.4,
        "fat": 2.8,
        "fiber": 4.1
      },
      "feeling": null,
      "feeling_time": null
    }
  ],
  "water_ml": 1200,
  "water_target_ml": 2100,
  "summary": {
    "calories": 1
...[truncated 2047 chars]

Remediation

View remediation

Remediation Suggestions

1. Make session-only handling the default and persist data only after explicit, field-specific consent.
2. Explain which fields will be stored, their destination, retention period, and deletion procedure before writing.
3. Minimize persisted data and omit cycle, allergy, and other health-related fields unless they are necessary for a requested feature.
4. Create ~/nutrition-data with owner-only permissions such as 0700, and create each record with permissions such as 0600.
5. Encrypt sensitive records using an OS-backed key store or another established secret-management facility; do not store encryption keys beside the records.
6. Use atomic file creation and updates to prevent temporary or partially written copies from receiving weaker permissions.
7. Define automatic retention limits and provide a command or workflow that securely deletes individual logs and the complete profile.
8. Warn users that filesystem backups and synchronization tools may copy the records beyond the original device.
9. Validate that storage paths remain inside the intended user-owned directory and do not follow attacker-controlled symbolic links.

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Natural-Language Policy Violations

Medium

• Category: Not specified by scanner
• Confidence: 93% confidence

Finding

The text explicitly states this is an English version of the skill, and the manifest name also includes an '-en' suffix, but the file does not offer users a language choice or explain a necessary locale restriction. That creates a natural-language locale policy concern under the rule for forced language without opt-in.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.