Security audit
Security checks for vulnerabilities and agentic risk
This skill is a plain academic literature search guide with no code execution, persistence, credential handling, or hidden data access.
This skill appears safe to install for academic literature searches. Users should know it may translate search terms into English for better database coverage and may activate on broad requests about common methods, markers, results, or figures in a research area.
Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.
Then synthesize across papers instead of writing a paper-by-paper summary unless the user asked for that.
## Output Rules
- Answer directly in chat unless the user asks for a file.
- Use inline citations such as PMID, PMCID, DOI, or direct links.
The instruction 'Use English search terms for database queries, even if the conversation is in Chinese' imposes a fixed language behavior rather than offering a user choice or documenting a justified locale constraint. This is a natural-language policy issue because it overrides the user's language context without opt-in.
The manifest says to use the skill when the user wants to know what studies in a field usually detect, report, or use, which is a wide class of common literature-analysis requests. While examples are provided later, the activation guidance does not include negative examples or explicit limits distinguishing when this skill should not be invoked versus a general search or summarization skill.
The example 'What do results sections in this field usually report?' is broad and could match many general synthesis or tutoring requests, not just this specialized deep-search workflow. The file does not pair these examples with negative examples or tighter trigger criteria to prevent unintended invocation.
The manifest describes activation via "$academic-deep-search" plus a general task like finding papers on a topic, but it does not define clear trigger boundaries, exclusions, or negative examples. This can make invocation scope ambiguous in contexts where users discuss academic searches generally.
The guidance says to "use English search terms for the query itself," which imposes a specific language policy in natural language. The file does not offer a language choice, opt-in, or explain a justified region- or database-specific exception.
No suspicious patterns detected.