🔀 协调员
v1.0.01.0.0 并行协调者角色。专职聚合多方 Agent 输出，减少主 Agent 单点负载。
0· 18·0 当前·0 累计
by 下载技能包
最后更新
2026/4/19
安全扫描
OpenClaw
安全
high confidenceThe skill is an instruction-only
评估建议
This skill is internally consistent: it simply instructs a coordinator to dispatch work to other Agents, wait for their completion, and produce structured summaries. Before installing, confirm your platform's session APIs (sessions_spawn / sessions_send) and knowledge-base access controls behave as expected: limit the coordinator's read/write permissions to only the required Agent outputs, ensure original reports cannot be silently modified, and log coordinator actions. Also verify authorization...详细分析 ▾
✓ 用途与能力
Name/description (a coordinator that aggregates multiple Agents) matches the SKILL.md instructions. The skill requests no binaries, env vars, or installs and only describes dispatching to other Agents, collecting outputs, and summarizing—exactly what a coordinator would need.
ℹ 指令范围
SKILL.md stays within coordinator duties (spawn sessions, collect task-completion events, deduplicate, summarize, reference original Agent submission times). It references platform actions like sessions_spawn/sessions_send and knowledge-base paths (knowledge-base/audit/...), which is expected but implies the coordinator needs read access to stored Agent outputs and the platform's session API. There are no instructions to read unrelated host files or environment variables. Caveat: the document relies on platform-level capabilities and does not specify authorization checks or explicit restrictions on how original reports are accessed or stored.
✓ 安装机制
Instruction-only skill with no install spec and no code files. This is low-risk and consistent with a role-playing/instructional coordinator skill.
✓ 凭证需求
No environment variables, credentials, or config paths are requested. The only external resources referenced are other Agents and knowledge-base file paths, which are proportionate to the stated purpose.
✓ 持久化与权限
always is false and disable-model-invocation is default (agent may invoke autonomously) which is normal. The skill does not request persistent system presence or modification of other skills' configs.
安全有层次，运行前请审查代码。
运行时依赖
🖥️ OSLinux · macOS · Windows
版本
latestv1.0.02026/4/19
Initial release of the coordinator skill. - Introduces a specialized coordinator role to aggregate outputs from multiple Agents, reducing main Agent load. - Defines four coordinator types (财务, 技术, 市场, 运营) and their aggregation logic. - Provides a structured workflow for collecting, sorting, summarizing, and forwarding reports to the CEO. - Outlines standardized output formats and coordinator activation rules. - Includes code example for coordinator assignment based on participating Agents. - Establishes clear operational rules: coordinator cannot make decisions, must aggregate faithfully, and always document source and timing.
● 无害
安装命令
点击复制官方npx clawhub@latest install coordinator-1-0-0
镜像加速npx clawhub@latest install coordinator-1-0-0 --registry https://cn.longxiaskill.com