Call Doubao1 5 Llm
PassAudited by VirusTotal on May 12, 2026.
Overview
Type: OpenClaw Skill Name: call-doubao1-5-llm Version: 0.1.0 The skill bundle contains standard metadata and prompt engineering instructions for an AI agent. The `SKILL.md` file uses strong language to enforce rules regarding output language, role-playing, and task focus, but these rules are aimed at ensuring predictable and aligned agent behavior, not at subverting the agent. There is no evidence of malicious intent, data exfiltration, unauthorized command execution, or attempts to bypass security measures. The instructions explicitly direct the agent to 'Stay Task-Focused' and 'Do not drift into topics unrelated to the current task'.
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.
The skill may strongly constrain how the agent responds, especially around language selection.
The skill uses absolute priority wording to control output language, which can steer the agent's behavior even though it is limited to response-language handling.
Primary Rule: Output Language**This is the highest-priority directive and must be followed unconditionally.**
Use this skill only when that language behavior is desired, and ensure normal user and platform instructions remain authoritative.
If the meeting minutes contain mistakes or embedded instructions, the agent may over-weight them when responding.
The skill directs the agent to rely heavily on meeting minutes as memory-like context, so inaccurate or manipulated minutes could shape the response.
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.
Provide trusted meeting minutes and treat any instructions inside them as context unless the user explicitly confirms them.
The skill may not behave like a general Doubao LLM-calling integration and may instead steer the agent into a creative meeting persona.
The actual prompt frames the skill as a creative brainstorming participant, while the registry description says it is for calling Doubao1.5 LLM tasks; this is a capability-description mismatch, but no hidden code or sensitive access is evidenced.
You are a creative expert and a core member of a top-tier creative team. Your codename is **$SHOW_NAME$**.
Review the prompt before installing and use it only if the creative brainstorming role matches your intended task.