Systematic Anatomy 10edition

• Outcome: Review
• Latest audit: Jun 5, 2026 at 2:25 PM UTC
• Version:
  1.0.0

Overview

This anatomy skill set is not executable malware, but it needs review because several modules give invasive or high-stakes medical procedure guidance without strong per-skill safety boundaries.

Install only if you want a medical anatomy reference for trained or supervised clinical education. Do not use it as a standalone source for diagnosis, treatment, injections, incisions, catheterization, sterilization procedures, or emergency care; those modules should be revised with clear clinician-only scope, red-flag escalation, and procedure-specific safety limits before broad use.

SkillSpector

By NVIDIA

Vulnerability Patterns

• 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
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (9)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger condition includes '其他下尿路梗阻症状', which is broad and can activate the skill for symptoms caused by conditions other than benign prostatic hyperplasia, such as urethral stricture, neurogenic bladder, infection, or malignancy. In a medical skill that discusses examination and causal interpretation, overly broad activation can misroute users into a narrowed BPH-focused workflow and increase the risk of inappropriate reassurance or delayed consideration of alternative diagnoses.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill gives procedural guidance for locating and using the great saphenous vein for infusion, blood draw, or injection, but it lacks explicit safety framing that this is an invasive clinical procedure requiring trained personnel, aseptic technique, and monitoring for complications. In a general-purpose agent skill library, that omission can enable unqualified users to attempt venipuncture and can lead to infection, arterial/nerve injury, extravasation, thrombosis, or delay of appropriate care.

Missing User Warnings

High

• Confidence: 96% confidence
• Finding: The skill gives actionable instructions for performing a cervical plexus anesthetic injection by identifying the nerve point and stating that local anesthetic can be injected there, but it provides no warning that this is an invasive medical procedure requiring trained clinicians, sterile technique, dosing safeguards, and monitoring for complications. In this context, omission of explicit medical-safety constraints is dangerous because cervical/neck injections carry meaningful risks including nerve injury, vascular injection, diaphragmatic/phrenic nerve effects, local anesthetic systemic toxicity, and airway-threatening complications.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill provides biomechanical assessment guidance and clinical interpretation such as explaining abnormal lumbar lordosis and linking findings to腰痛、姿势代偿或脊柱变形风险, but it does not tell users this is informational only or not a substitute for professional evaluation. In a medical context, users may over-rely on the output for self-assessment or rehabilitation decisions, increasing the risk of delayed diagnosis or inappropriate self-management.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger condition includes broad clinical contexts such as female reproductive diagnosis or surgical planning, which can cause the skill to activate outside its stated narrow purpose of fallopian tube segment anatomy. In a medical setting, over-broad activation can surface partial or context-inappropriate guidance and may influence sensitive diagnostic or procedural reasoning without sufficient safeguards or specialization boundaries.

Missing User Warnings

High

• Confidence: 95% confidence
• Finding: The skill gives operational guidance for planning and executing a surgical skin incision, but it does not include any warning that this is a medical procedure requiring licensed clinical judgment, patient-specific assessment, sterility, and supervision. In an agent setting, this creates a meaningful risk that non-qualified users could treat the instructions as actionable procedural advice, leading to patient harm, poor wound healing, infection, or scarring.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: This skill provides clinical localization and diagnostic inference for spinal cord injury patterns and explicitly says the output is used to guide imaging, treatment decisions, and prognosis, but it does not clearly warn users that it must not be used as the sole basis for diagnosis or treatment. In a high-stakes medical context, omission of that limitation can lead users to over-trust the heuristic and make unsafe triage or management decisions without appropriate specialist evaluation or confirmatory testing.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: This skill provides procedural guidance for a sterilization-related surgery and even recommends a preferred ligation site, but it omits explicit warnings that vasectomy is intended as a permanent contraceptive procedure and must be performed only by qualified clinicians with informed consent and patient-specific judgment. In an agent setting, such omission can normalize or facilitate unsafe medical action by unqualified users, increasing the risk of irreversible harm, complications, and misuse.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill includes clinically consequential guidance about pericardial fluid distribution, pericardiocentesis target selection, and surgical clamping anatomy, but it does not clearly restrict use to trained medical professionals or warn against lay execution. Because this is procedural medical content, omission of an explicit safety boundary can enable misuse or overreliance in situations where anatomical variation, pathology, or urgency make simplified guidance unsafe.

VirusTotal

VirusTotal findings are pending for this skill version.