Security audit
Security checks for vulnerabilities and agentic risk
This textbook-style medical cell biology skill set includes step-by-step cloning, wet-lab, and cancer-treatment guidance that needs review before use.
Install only if you want an educational/research reference and can enforce your own medical, biosafety, legal, and institutional review boundaries. Do not use it as standalone clinical advice or as a protocol source for embryo work, cloning, hazardous chemistry, radioactive assays, or cell-culture engineering.
This section gives actionable steps for therapeutic and reproductive cloning, including embryo culture to blastocyst, embryonic stem-cell derivation, and implantation into a surrogate. Such operational guidance can facilitate highly regulated and potentially unlawful cloning activity, and the medical-textbook context does not justify providing execution-ready instructions.
The skill explicitly recommends cell immortalization methods such as radiation exposure, chemical carcinogen treatment, and oncogene transfection, but provides no safety framing, access controls, or restriction to qualified biosafety contexts. These are hazardous wet-lab interventions that can enable creation or propagation of transformed cell lines, so presenting them as straightforward options materially lowers the barrier to unsafe experimentation.
A skill described as executing somatic cell nuclear transfer and reprogramming for cloning or pluripotent stem cell generation materially exceeds a passive educational catalog and moves into actionable wet-lab bioengineering guidance. In this context, the medical education framing makes it more concerning, because advanced experimental procedures with ethical, biosafety, and regulatory implications are embedded in a broad learning package without visible access controls, safeguards, or restrictions.
The skill provides step-by-step instructions for a hazardous wet-lab setup involving flammable gases (CH₄, H₂), toxic/corrosive ammonia, vacuum operations, boiling liquids, and a 60 kV spark discharge, but includes no safety warnings, qualification requirements, PPE guidance, ventilation requirements, or supervision constraints. In a general-purpose agent skill, this omission materially increases the risk that an unqualified user could attempt the procedure and cause fire, explosion, electric shock, glass implosion, or inhalation injury.
The skill goes beyond educational cell-biology content and provides an operational SCNT workflow culminating in reproductive cloning for nonhuman use. Even with a narrow disclaimer against human reproductive cloning, the instructions meaningfully enable embryo manipulation and cloning procedures that are highly sensitive, dual-use, and misaligned with the stated textbook-style educational purpose.
The skill describes creation of patient-specific embryonic stem cells and cloning workflows with only a limited prohibition on human reproductive cloning, but without robust warnings about embryo research restrictions, biosafety controls, consent, animal welfare, and jurisdiction-dependent illegality. That omission increases the risk that users interpret the content as acceptable operational guidance rather than tightly restricted subject matter.
The skill gives concrete treatment-selection and drug-use guidance for high-risk oncology care, including named agents and implied administration, without any safety disclaimer, clinician-only restriction, emergency warning, or instruction to defer to current specialist protocols. Because cancer therapy decisions are life-critical and highly patient-specific, a user could over-trust this content and apply incomplete or outdated recommendations, causing serious harm.
Suspicious Unicode normalization or mixed-script content
The line claiming users can 'just input one question' and automatically get a solution from the textbook overstates the invocation scope and encourages use for arbitrary prompts without clear boundaries. In an agent setting, overly broad triggering can cause the skill to activate outside its intended educational domain, increasing the chance of unsafe medical-style guidance or low-relevance responses being presented with undue authority.
Suspicious Unicode normalization or mixed-script content
Suspicious Unicode normalization or mixed-script content
Suspicious Unicode normalization or mixed-script content
Suspicious Unicode normalization or mixed-script content
The trigger description is broad enough to match many drug transport, metabolism, and chemotherapy-resistance discussions without clear activation boundaries. This can cause the skill to be invoked in contexts where it is only partially relevant, increasing the chance of misleading analysis or overconfident medical guidance, though it does not directly create code-execution or prompt-injection risk.
This file contains user-facing natural-language instructions entirely in Chinese, including the name, description, trigger conditions, and usage notes. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly documented and justified, which is not present here.
The file forces a specific language for the skill’s description and operating instructions, which can violate language/locale policy when no user opt-in or alternative is provided. There is no indication that this skill is region-specific or that Chinese is required for compliance or domain reasons.
Suspicious Unicode normalization or mixed-script content
This file presents the skill name, description, and usage instructions exclusively in Chinese. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is clearly documented and justified, which is not present here.
Suspicious Unicode normalization or mixed-script content
This markdown skill enforces a specific language/locale for its name, description, and usage guidance. Under the policy, forcing a language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly documented and justified, which is not present here.
The skill name, description, headings, and output text are entirely in Chinese, which effectively imposes a specific language/locale on users. The file does not indicate that Chinese is optional, user-selected, or required for a region-specific purpose, so this is a natural-language policy issue under the locale-choice rule.
This file forces a specific language/locale in its natural-language interface: both the manifest description and all user-facing guidance are exclusively in Chinese. Under the policy, language constraints should not be imposed unless the skill offers user choice or clearly documents a justified region-specific limitation.
The manifest name, description, headings, and procedural content are entirely written in Chinese, which imposes a specific language on users without any stated opt-in or alternative. Under the policy, language constraints should either be optional for the user or clearly documented as necessary for a region- or locale-specific use case.
The skill name, description, and operating instructions are entirely in Chinese, with no indication that language selection is optional or that the skill is intentionally limited to Chinese-speaking users. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.
The file’s name, description, usage guidance, and procedural steps are all presented only in Chinese. Under the policy for natural-language violations, forcing a specific language without user opt-in should be flagged unless the locale restriction is explicitly justified, which is not present here.
No suspicious patterns detected.
