Back to skill

Security audit

Hebrew

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only Hebrew writing style guide with no code or system access; users should just know it strongly defaults to casual Israeli Hebrew.

Install this if you want Hebrew output to sound casual and Israeli. For formal documents, neutral Hebrew, or another Hebrew variety, explicitly state that preference in your prompt.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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 (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The instruction 'Unless explicitly formal: lean casual' sets a mandatory default register tied to Israeli Hebrew norms rather than offering the user a choice. This is a natural-language policy concern because it forces a locale/style preference unless the user explicitly overrides it.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The line 'Israeli culture is very informal' is used as prescriptive guidance for output style, steering responses toward one locale-specific norm. Because the file does not provide opt-in, alternatives, or a scope limitation, it risks imposing a particular regional language convention on all users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The 'Native Test' says to add 'sabra flavor' if the text seems AI-generated, which prescribes a specific Israeli cultural-linguistic style. This can override user expectations for other Hebrew varieties or more neutral Hebrew because no user opt-in or locale choice is offered.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.