Security audit

Meta Analysis Topic Finder

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent research workflow for assessing meta-analysis topics and generating local reports, with no evidence of hidden access, persistence, or unsafe execution.

Before installing, understand that the skill may guide web/database searches and create local report files from your research inputs. Use project-specific output paths, review any generated methodology recommendations, and treat rapid assessments as tentative until deduplication searches are completed.

A.I.G

Vulnerability Patterns

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
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (10)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The trigger phrase "Tell me in 5 min whether this can be a meta" is broad natural language that could plausibly appear in general conversation about research help, causing the skill to activate when the user did not explicitly request this workflow. Over-broad activation can route users into an unintended skill, leading to context hijacking, incorrect handling, or interference with more appropriate tools.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 89% confidence
Finding: The skill instructs the agent to read multiple local reference files and generate/save report files, but it does not declare any explicit tool permissions or allowed-tools scope. That mismatch creates an authorization ambiguity: a host agent may grant broader file read/write capability than users expect, enabling unintended access to local files or creation/overwrite of files during execution.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The template begins with a Chinese title field ({{title_zh}}) and then labels English as a secondary field, which implies a required language ordering. Because the file does not state that the skill is intended only for Chinese-language users or provide an opt-in language choice, this appears to impose a locale preference in natural-language content.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 84% confidence
Finding: Lines L169-L173 state that users 'must' extend searches to non-English databases whenever certain population, market, or publication-language conditions apply. This imposes a language/locale-specific requirement as policy, but the document does not frame it as optional user choice or explicitly justify it as a region-specific compliance requirement.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The example report hard-codes the search language scope to English + Chinese without documenting that this was user-selected or methodologically justified. In a meta-analysis topic-selection skill, this can systematically bias evidence identification, omit relevant studies from other languages, and produce misleading feasibility, novelty, or downstream review recommendations.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 87% confidence
Finding: This markdown text describes the skill as having been fully translated to English for all files, which can imply a fixed language policy. Under the policy rule, forcing a specific language without user opt-in may be a locale-policy violation, and no opt-in or region-specific justification is provided here.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 82% confidence
Finding: Line L077 instructs the user to cite a fixed English sentence: "This systematic review is registered with PROSPERO...". Because the file otherwise discusses non-English titles and multilingual registrations, prescribing a single language string without opt-in or locale guidance is a natural-language locale policy concern.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The generated HTML sets <html lang="en"> unconditionally even though the report content and example data include Chinese fields such as title_zh and Chinese-language database content. This forces a specific language/locale in output metadata without offering the user a choice or documenting that the tool is English-only.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 78% confidence
Finding: Line L246 states a fixed language scope of 'English + Chinese'. Under the policy rule, forcing a specific language or locale without opt-in or clear justification can be a natural-language policy issue; this file does not explain why only those languages are included or offer any alternative.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The document states a fixed language scope of "English + Chinese," which imposes a locale/language constraint in the workflow. There is no accompanying opt-in, user choice, or explicit justification that the review is intentionally limited to those languages for a documented regional or methodological reason.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.