Security audit

FitClaw Public Core

Security checks for vulnerabilities and agentic risk

Overview

This fitness coaching skill is coherent, but it asks agents to remember health-adjacent profile and behavior data without clear consent, retention, or deletion controls.

Review this skill before installing if you do not want fitness, body, nutrition, or habit information retained in agent memory. It should ideally be updated to ask before saving health-adjacent details, store only what is necessary, and tell users how remembered data can be reviewed or deleted.

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

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
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (4)

Natural-Language Policy Violations

Medium

Confidence: 81% confidence
Finding: The skill gives sex-specific carbohydrate targets for men and women as a default rule without offering a neutral alternative, user choice, or justification for why sex is required. In a nutrition coaching context, this can lead to exclusionary or inaccurate recommendations for users who do not fit binary categories, and may encourage collection or inference of sensitive sex/gender data that is not strictly necessary.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill explicitly recommends storing user memory items including goal direction, nutrition structure, recurring food behavior problems, and adherence trends, which can constitute sensitive health-related profile data. In a public-safe coaching module, omitting retention notice, minimization limits, consent, and deletion guidance increases privacy risk and creates a path for unnecessary collection or persistence of sensitive user information.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill explicitly recommends storing user training preferences, strength anchors, and recent feedback in memory, but provides no user-facing notice, consent boundary, retention limit, or data minimization guidance. In a fitness coaching context, this can create privacy risk because behavioral and health-adjacent data may be retained or reused without the user's awareness, especially across sessions or platforms.

Missing User Warnings

Low

Confidence: 89% confidence
Finding: The skill explicitly recommends storing hydration targets, reminder preferences, and adherence trends in user memory, but it does not mention retention limits, consent, deletion, or privacy handling. While the data is relatively low sensitivity in this public-safe fitness context, silent persistence of behavioral/health-adjacent information can still create privacy risk and unexpected data retention.

Static analysis

No suspicious patterns detected.