Surgery-PMPH-10edition

• Outcome: Review
• Latest audit: Sep 16, 2026 · 9:20 AM UTC
• Version: 1.0.1

Overview

The skill is not malware, but it needs review because it contains many high-risk surgical and emergency medical instructions with weak safety boundaries.

Install only if you want a clinical education/reference skill and will treat it as decision support for qualified medical review, not as patient self-care or autonomous treatment guidance. For emergencies, invasive procedures, medication dosing, surgery, anesthesia, resuscitation, oncology, transplant care, or serious infection, rely on licensed clinicians, local protocols, and urgent in-person care.

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

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

Findings (338)

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

This README encourages users to 'just ask a question and get a solution directly from the textbook' and later instructs the agent to verify a medical skill before every answer, but it does not clearly warn that outputs are educational only and not a substitute for clinician judgment or patient-specific advice. In a medical skill, that omission materially increases the risk that users or downstream agents rely on generated content for diagnosis or treatment decisions without appropriate safeguards.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

This is a true safety vulnerability: the skill gives step-by-step guidance for invasive, high-risk medical interventions such as bladder pressure measurement, mechanical ventilation, percutaneous drainage, and decompressive laparotomy, but it does not clearly warn that these actions are only for qualified clinicians in appropriate care settings. If surfaced to a general user or used without specialist judgment, the content could enable dangerous self-treatment or inappropriate clinical action, causing severe injury or death.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

The skill provides clinician-grade medical instructions for an invasive diagnostic procedure (graft biopsy) and high-risk rescue treatments including pulse-dose methylprednisolone, ATG/ALG, plasmapheresis, and immunosuppression escalation, but it does not include an explicit clinician-only restriction, emergency referral language, or patient safety warning. If surfaced to a lay user or applied without specialist oversight, it could prompt dangerous self-management or inappropriate treatment decisions, leading to severe harm, graft loss, infection, or death.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

This skill provides actionable burn shock fluid resuscitation instructions, including formulas, fluid composition, infusion timing, and route recommendations, but does not explicitly warn that it is not a substitute for clinician judgment or urgent emergency care. In a high-risk acute care scenario, users could over-rely on the skill and apply outdated, incomplete, or context-inappropriate guidance, causing under-resuscitation, fluid overload, electrolyte errors, or delayed transfer to emergency treatment.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 99% confidence

Finding

This skill gives step-by-step instructions for a hazardous orthopedic reduction maneuver but does not explicitly warn that it is only for trained clinicians in an emergency/operative setting with anesthesia, imaging, and post-reduction neurovascular assessment. Procedural medical guidance without clear role restriction and supervision language increases the risk that an unqualified user may attempt a dangerous intervention, causing nerve, vascular, or additional joint injury.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

This skill provides individualized antibiotic regimen design guidance, including route, dose, frequency, duration, and combination therapy, but contains no warning that the output is high-risk medical information requiring qualified clinician review. Because inappropriate antibiotic selection or dosing can cause patient harm, treatment failure, toxicity, or antimicrobial resistance, omission of safety framing and clinical-oversight requirements is dangerous in this context.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

This skill gives prescriptive medical treatment instructions for suspected or confirmed bone and joint tuberculosis, including when to initiate therapy, drug combinations, duration, adverse-effect management, and cure criteria, without a clear warning that it is informational only and must be used under qualified clinician oversight. Because anti-tuberculosis treatment is high-risk, prolonged, and requires individualized diagnosis, dosing, lab monitoring, drug-resistance assessment, and management of toxicity, an autonomous agent or lay user could misuse the content and cause serious harm through delayed care, inappropriate treatment, hepatotoxicity, resistance, or treatment failure.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill provides concrete hemodynamic treatment directives, including rapid fluid resuscitation and named vasoactive/cardiac medications, but does not clearly state that the guidance is not a substitute for clinician judgment or that misuse can cause serious harm. In a clinical decision-support context, terse algorithmic instructions can be over-trusted, increasing the risk of inappropriate treatment, fluid overload, cardiogenic deterioration, medication misuse, or delayed recognition of alternative shock states.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill provides concrete cancer subtype classification and treatment recommendations, including endocrine, chemotherapy, HER2-targeted, and immunotherapy options, without an explicit warning that the output is informational only and must not replace oncologist judgment. In a high-stakes medical context, users may over-trust the advice despite omitted clinical factors such as stage, comorbidities, pathology details, contraindications, pregnancy status, and current guideline variations, creating risk of harmful self-management or inappropriate clinical use.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 96% confidence

Finding

This skill provides clinical burn-depth classification and treatment-linked guidance, but it does not warn users that the content is not a substitute for professional diagnosis or emergency care. Because burn severity can be misjudged early and incorrect depth assessment can delay urgent treatment, fluid resuscitation, referral, or surgery, a user may over-rely on the skill in a time-sensitive medical situation.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

This skill provides step-by-step instructions for a manual fracture reduction and cast immobilization procedure, which is an invasive clinical intervention that can cause neurovascular injury, worsened displacement, compartment syndrome, or improper healing if performed by an unqualified user. The content lacks explicit warnings that it is for trained clinicians only, does not require imaging/clinical assessment safeguards, and omits urgent escalation criteria, making misuse more likely and more dangerous.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill gives step-by-step instructions for invasive procedures involving abdominal puncture and lavage, including anatomical landmarks, instrument use, fluid volumes, and interpretation criteria, but does not explicitly state that it is only for trained clinicians and not for self-use. Because these procedures can cause hemorrhage, bowel perforation, infection, and delayed definitive care if attempted by unqualified users, the omission materially increases misuse risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill provides step-by-step selection and use of invasive diagnostic and therapeutic endoscopic procedures, including EMR/ESD, dilation, stenting, ERCP, EST, hemostasis, and stone extraction, but does not clearly state that these are specialist-only procedures requiring licensed gastroenterologists, facility support, informed consent, monitoring, and emergency backup. In a generic agent setting, this omission is dangerous because it can normalize or operationalize high-risk medical procedures without adequate warnings about perforation, bleeding, pancreatitis, sedation risk, infection, or the need for formal clinical evaluation.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill provides step-by-step instructions for an invasive surgical incision-and-drainage procedure, including anesthesia, incision placement, tissue dissection, and debridement, but does not clearly warn that this is a high-risk medical procedure requiring qualified clinicians, sterile technique, and appropriate facilities. In a general-purpose agent skill context, this omission materially increases the chance of unsafe use by untrained users, leading to infection spread, neurovascular injury, ischemia, permanent functional loss, or delayed definitive care.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

The skill gives step-by-step guidance for manual closed reduction of a dual-bone forearm fracture, which is a high-risk medical procedure requiring trained clinical judgment, imaging interpretation, and neurovascular reassessment. Presenting procedural instructions without clear warnings to seek urgent professional care or limiting use to qualified clinicians could enable unsafe self-treatment or unqualified intervention, risking malreduction, neurovascular injury, compartment syndrome, and permanent disability.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 96% confidence

Finding

The skill gives concrete diagnostic differentiation and drug treatment instructions, including specific antibiotics, dosing, and escalation advice, without any warning that the content is informational only and must not replace clinician assessment. In a medical STI context, users may over-trust the skill and act without confirmatory testing, allergy review, resistance awareness, or evaluation for co-infections and contraindications, creating a real risk of patient harm.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

This skill provides step-by-step guidance for emergency invasive liver trauma surgery, including laparotomy, vascular inflow occlusion, and vena cava control, without any explicit warning that it is for trained clinicians only and unsuitable for lay execution. In a general-purpose agent setting, omission of strong safety framing materially increases the risk of severe injury or death if the instructions are surfaced to an unqualified user or used without urgent specialist oversight.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 96% confidence

Finding

The skill directs an immediate irreversible medical action—removal of a transplanted organ—based on pattern-matched symptoms, but provides no requirement for urgent specialist escalation, confirmation safeguards, or user-facing warning that this is high-risk clinical guidance. In a medical agent context, presenting destructive treatment steps without strong boundary checks could cause catastrophic patient harm if applied to a misdiagnosed or non-clinical scenario.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

This index advertises and links to a large library of high-risk clinical and emergency intervention skills, including airway management, CPR, defibrillation, fluid resuscitation, emergency surgery, anesthesia, and drug use, but it does not prominently warn that the content is for trained medical professionals and not for layperson self-treatment. In a general-purpose agent context, the absence of clear safety boundaries increases the chance that unsafe users may treat the material as direct actionable guidance, leading to severe patient harm or death.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill provides diagnostic and treatment guidance for a serious intra-abdominal infection, including antibiotics, postoperative management, laparotomy, and image-guided drainage, without any warning that this is for licensed clinicians or that urgent in-person medical evaluation is required. In an agent setting, this could be surfaced directly to lay users and may prompt delayed emergency care, self-management, or unsafe interpretation of surgical indications.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 96% confidence

Finding

The skill provides stepwise guidance for invasive abscess drainage and incision procedures, including puncture, catheter placement, rectal/vaginal drainage, and postoperative management, but does not clearly restrict use to licensed clinicians or warn that misuse can cause serious patient harm. In a general-purpose agent environment, this omission increases the risk that unqualified users may attempt or over-trust high-risk medical procedures, leading to bleeding, perforation, delayed definitive care, or infection-related deterioration.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

The skill gives step-by-step instructions for an invasive emergency medical procedure, including pressures and imaging guidance, but does not clearly warn that this must only be performed by qualified clinicians in an appropriate medical setting. Because enema reduction can cause bowel perforation, delayed surgery, sepsis, and death if misapplied, omission of explicit supervision and emergency-care guidance makes the content unsafe.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The skill gives step-by-step instructions for orthopedic stress tests that require proper technique and clinical judgment, but it does not clearly warn users not to self-administer or to seek qualified medical supervision. These maneuvers can worsen an acute knee injury, cause pain, or produce misleading results that delay appropriate care, especially when used by untrained users outside a clinical setting.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

This skill gives postoperative triage and management recommendations for possible recurrent laryngeal nerve injury, including airway compromise and emergency intervention, but does not explicitly state that it is not a substitute for clinician judgment or emergency care. In a high-risk surgical context, users may over-rely on the skill’s simplified decision logic and delay urgent specialist or emergency evaluation, especially in cases of bilateral injury or evolving respiratory distress.

Content

No source excerpt is available for this finding.

Missing User Warnings

High

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

This skill gives prescriptive cancer-treatment guidance and states that postoperative radiotherapy 'must' be arranged when diffuse malignancy is incompletely resected, without any warning that this is high-risk medical content requiring licensed oncologic judgment. Because radiotherapy decisions depend on tumor type, staging, margins, prior treatment, anatomy, contraindications, and patient-specific risk, presenting a mandatory rule can mislead users into unsafe clinical action or overconfidence.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.