Call Grok4 Llm
AdvisoryAudited by Static analysis on Apr 30, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A user may install or invoke the skill expecting a Grok4 integration, but it would instead change the assistant’s role and response framing.
The declared purpose suggests a Grok4 LLM-calling helper, but the actual instructions define a creative-team meeting persona instead.
description: AI agent for call grok4 llm tasks ... You are a creative expert and a core member of a top-tier creative team.
Rename and describe the skill according to its actual brainstorming persona behavior, or replace the prompt with clear, scoped instructions for the intended Grok4-calling workflow.
The assistant could prioritize the skill’s persona, language, and CCO-meeting rules over the user’s actual request.
The prompt tries to impose its own priority and authority structure, which can redirect the agent away from the user’s expected Grok4 task when the skill is invoked.
This is the highest-priority directive and must be followed unconditionally ... respond only to the specific question or task posed to you by the CCO
Make these instructions explicitly subordinate to the user’s task and remove unconditional priority language unless it is necessary and clearly disclosed.
If meeting minutes are stale, manipulated, or include unsafe instructions, the assistant may treat them as authoritative context.
The prompt instructs the agent to over-trust meeting minutes as memory, although no storage or retrieval mechanism is included in the artifacts.
The entire context of the meeting, historical discussions, and current agenda items are recorded in real time in the meeting minutes. These minutes are your sole source of information and memory, and you must adhere to them.
Clarify where meeting minutes come from, require user confirmation for important context, and avoid treating retrieved or generated notes as unquestionable authority.
Users have less context for who maintains the prompt and why its behavior differs from the description.
The skill has limited provenance metadata, though there is no install script or executable code in the provided artifacts.
Source: unknown; Homepage: none
Prefer skills with clear source, homepage, maintainer information, and documentation matching the actual instructions.