Security audit

wikipedia-oc

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward English Wikipedia lookup helper with disclosed limitations and no hidden persistence or destructive behavior.

Before installing, understand that this skill retrieves from English Wikipedia even when your question is in another language, so translated answers may miss details from local-language articles. Install the Python dependency in a virtual environment and consider pinning a reviewed version of the wikipedia package.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
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

Findings (1)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:35
Finding: Unpinned Third-Party Python Dependency
Content: View full analysis
Remediation: View remediation
" ``` - Record and verify cryptographic hashes in a requirements or lock file, and install with hash enforcement: ```bash python -m pip install --require-hashes -r requirements.txt ``` - Lock and review all transitive dependencies, not only the direct `wikipedia` dependency. - Obtain packages exclusively from an explicitly configured trusted index. - Prefer reviewed binary wheels where appropriate and prohibit unexpected source builds. - Install the dependency in an isolated virtual environment using an unprivileged account. - Integrate dependency vulnerability and provenance checks into the release process, and update the pinned version only after review and testing. ]]>

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill hard-codes English-only retrieval even when the user asks for another language, which can silently override user intent and lead to mistranslation, omission of locale-specific information, or incorrect answers presented as if they satisfy the original request. In a retrieval skill, forcing a different source language without explicit consent is a policy and integrity risk because the post-processing translation step may distort meaning while hiding that the original lookup was not performed in the requested language.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.