Security audit

Call Grok4 Llm

Security checks for vulnerabilities and agentic risk

Overview

The package claims to be a Grok 4 calling skill, but its only artifact instead rewrites the agent’s role and instruction hierarchy.

Review before installing. This skill does not appear to implement a transparent Grok 4 integration; it mainly changes how the agent should behave, who it should act as, and what context it should obey. Ask the publisher to replace the role-hierarchy prompt with narrow Grok 4 invocation instructions, clear input/output boundaries, and explicit handling of untrusted meeting-minutes content.

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

Findings (1)

T01 · Skill Instruction Hijacking

Error

Location: SKILL.md:13
Finding: Skill-Level Instruction and Role Hijacking
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (1)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill description, 'AI agent for call grok4 llm tasks,' is so broad that it does not meaningfully constrain when the skill should be invoked or what operations it is authorized to perform. In an agentic environment, this ambiguity can cause over-invocation, unintended delegation, and misuse in unrelated contexts, increasing the risk of prompt-injection exposure or unsafe handling of sensitive data.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.