菜单
Use when work needs persistent multi-step execution, independent review, resumable project state, or long-running research and engineering coordination through Argus.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
基于 SOC 职业分类
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/lbx154/Argus --skill argus-run命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
Plan and execute multi-step research, surveys, feasibility studies, and evidence-heavy analyses with adaptive planning, source verification, critic review, and cross-task learning. Use when a request needs multiple dependent investigations or a durable evidence trail; do not use for simple factual questions or small one-step edits.
Use when checking an Argus project's progress, pending questions, active role, diagnostics, selected artifacts, intervention state, recovery options, or graceful stop status.
Use when researching a biomedical target and disease relationship, mechanism, human evidence, clinical translation, safety, failed programs, competitive trials, or an auditable pharmaceutical decision dossier.
| name | argus-run |
| description | Use when work needs persistent multi-step execution, independent review, resumable project state, or long-running research and engineering coordination through Argus. |
Use the installed Argus MCP tools. Do not imitate the Argus roles in the host session and do not write Argus state files directly.
argus_project_list with the exact current work directory.argus_project_create with that directory and a concise name.argus_message with the user's complete objective. Manager owns chat
versus task routing, lifetime, workflow, domain, and execution handoff.argus-status Skill when status is requested.If the MCP server is unavailable, run argus --doctor and report the exact
installation or backend failure. Do not silently replace Argus with an ad hoc
single-agent workflow.
Never claim completion from dispatch, daemon activity, loss, or intermediate artifacts. Argus completion comes from its persisted Reviewer/Manager state.