Security audit

Caregiving Physical Skills

Security checks for vulnerabilities and agentic risk

Overview

This is a caregiving guidance skill with disclosed health and reminder content, but users should review its medical boundaries and installation permissions before use.

Install only if you are comfortable with a caregiving skill that stores or uses sensitive care context for reminders, and prefer a host setup that restricts filesystem access to a skill-specific state directory. Treat the physical-care and medication sections as educational support, not medical judgment; use clinicians, pharmacists, emergency services, and trained in-person help for unsafe transfers, medication changes, acute symptoms, serious falls, or wounds.

A.I.G

Vulnerability Patterns

Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Insecure DependenciesIntroduces malicious components through unsafe dependency sources
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

Findings (2)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:15
Finding: Unpinned npx Package Execution Creates a Supply-Chain Risk
Content: View full analysis
Remediation: View remediation

T05 · Unauthorized Access and Privilege Escalation

Note

Location: SKILL.md:13
Finding: Filesystem Capability Exceeds the Skill's Documented Operational Requirements
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (6)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The top-level description is broad enough that the skill may activate in many everyday caregiving situations without clear limits on when professional medical escalation is required. In a health-related skill, overbroad triggering is more dangerous because users may receive highly specific physical-care instructions in situations that exceed safe home-care boundaries.
Content: No source excerpt is available for this finding.

Rp1

Medium

Category: MCP Rug Pull
Confidence: 86% confidence
Finding: The manifest includes an install command using npx clawhub without a pinned version or integrity constraint. This creates a supply-chain risk because installation behavior can change over time or be influenced by a compromised upstream package, causing users to fetch and execute unexpected code during install.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The 'When to Use' section lists very broad scenarios, including medication management and warning-sign recognition, without concrete constraints or disqualifiers. In this caregiving context, that can lead to the skill being invoked for complex medical or physical-assistance situations where generalized instructions are insufficient and could delay emergency care or contribute to injury.
Content: No source excerpt is available for this finding.

Autonomous Decision Making

Medium

Category: Excessive Agency
Confidence: 75% confidence
Finding: Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
Content: Scanner excerpt · SKILL.md (reported line 292)May include surrounding context.
md
dangerous interactions that individual doctors might miss. CRITICAL RULES: - NEVER crush or split a pill without checking with the pharmacist. Some pills are extended-release or enteric-coated — crushing them dumps the full dose at once, which can be fatal. - If they can't swallow pills, ask the pharmacist about liquid

Description-Behavior Mismatch

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The manifest frames the skill as focused on physical caregiving techniques such as transfers, bathing, and movement assistance. The documented scope and detailed instructions also cover managing another person's medications, which is adjacent caregiving guidance rather than a physical care technique.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Low

Category: Not specified by scanner
Confidence: 82% confidence
Finding: The skill's stated purpose emphasizes hands-on physical care skills for caregivers. It extends beyond that into recognizing stroke, dehydration, UTI, and silent heart attack warning signs, which is health-triage education rather than a physical technique.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.