什么是 Social Ops？

social-ops 是一个为 AI 代理设计的复杂框架，用于处理社交媒体运营的全生命周期。通过将复杂的社交任务分解为六个不同的角色——侦查员（Scout）、研究员（Researcher）、内容专家（Content Specialist）、响应者（Responder）、发布者（Poster）和分析师（Analyst），它确保了每一项行动都是专注的、可审计的，并符合战略目标。该技能在管理 Moltbook 等平台上的大批量互动时特别有效，在这些平台上，一致性和品牌语调至关重要。

这套 Openclaw 技能集合构建为可组合且可移植的，允许开发人员编排从最初机会探测到最终绩效分析的无缝流水线。它将“思考”与“执行”分离，确保内容在发布给现场受众之前经过研究和审查。

下载入口:https://github.com/openclaw/skills/tree/main/skills/dougbtv/social-ops

安装与下载

1. ClawHub CLI

从源直接安装技能的最快方式。

2. 手动安装

将技能文件夹复制到以下位置之一

优先级：工作区 > 本地 > 内置

3. 提示词安装

将此提示词复制到 OpenClaw 即可自动安装。

Social Ops 应用场景

• 自动化发现热门趋势和 submolts 以进行早期参与。
• 构建具有特定消息路径和节奏的结构化内容流水线。
• 通过符合品牌语调的自动化回复来扩展社区管理。
• 管理高频发布计划，同时保持严格的审计日志。
• 利用分析师角色收集的数据驱动洞察来迭代社交策略。

1. 侦查员监控社交环境以发现新兴机会或相关对话。
2. 研究员对确定的主题进行深入研究，提供背景信息和指导。
3. 内容专家将研究和策略转化为具体的帖子和路径积压工作。
4. 响应者或作者起草特定的内容或回复，匹配既定的品牌语调。
5. 发布者执行已完成项目的发布，并将其移入永久的“已完成”日志。
6. 分析师审查绩效产物，为下一个周期推荐策略调整。

Social Ops 配置指南

要初始化 social-ops 技能，必须为数据目录设置环境变量。这可以确保代理知道在哪里存储日志、队列和记忆。将以下内容添加到您的 shell 配置文件中：

export SOCIAL_OPS_DATA_DIR=/path/to/Social

定义目录后，确保将策略文档放置在 {baseDir}/assets/strategy/ 中，以便各角色参考北极星目标。您还可以通过运行提供的 cron 安装脚本来部署自动化的 Openclaw 技能工作流：

./packaged-scripts/install-cron-jobs.sh

Social Ops 数据架构与分类体系

该技能通过严格的目录契约来组织其操作，以保持可移植性：

name: social-ops
description: >
  Role-based social media operations skill. Use this skill when executing
  structured social campaigns — scouting opportunities, crafting content,
  posting, responding to threads, and analyzing performance. Designed for
  Moltbook engagement but adaptable to any platform and persona. Separates execution into six composable roles (Scout, Researcher,
  Content Specialist, Responder, Poster, Analyst) so each run stays focused
  and auditable. Activate this skill for any social ops task: opportunity
  detection, content pipeline management, community engagement, or
  performance review.

Social Ops

Execute social media operations through specialized roles. Each role has a single responsibility, reads its own reference doc, and hands off to the next stage in the pipeline.

NOTE: Utilize the moltbook skill as necessary to meet these goals.

Workflow

Scout ──→ Content Specialist  (new opportunities → lane strategy)
Scout ──→ Responder           (reply-worthy threads → responses)
Researcher ──→ guidance for Content Specialist & Writer
Content Specialist ──→ Writer (lanes → final posts)
Writer ──→ Poster             (finished posts → published)
Poster ──→ done logs          (published → archived)
Analyst ──→ strategy adjustments (performance data → tuning)

Roles

When dispatched to a role, read its reference doc fully before acting.

Dispatching a Role

1. Identify which role the task requires.
2. Read the full role doc at {baseDir}/references/roles/.md.
3. Follow the role's instructions — stay within its scope.
4. Hand off outputs to the next role in the workflow.

Strategy

The north-star strategy lives at {baseDir}/assets/strategy/Social-Networking-Plan.md. Read it before any Content Specialist or Analyst run. It defines brand voice, target audience, lane structure, and growth objectives.

Role I/O Map

Role-to-role artifact flow and logging ownership are documented in:

• {baseDir}/references/ROLE-IO-MAP.md

Environment Variables

Setup

Before any role can run, SOCIAL_OPS_DATA_DIR must be set. If it is not set:

1. Ask the operator where their Social data directory lives.
2. Recommend they add it to their shell profile:

export SOCIAL_OPS_DATA_DIR=/path/to/Social

All role docs reference $SOCIAL_OPS_DATA_DIR/ as the root for runtime data. This replaces the previous /Social/ convention for reliability.

Path Conventions

Use these path rules to keep the skill portable:

• Skill-owned files (docs, scripts, assets): use {baseDir}/...
• Runtime/social data files (logs, guidance, todo/done queues): use $SOCIAL_OPS_DATA_DIR/...
• Runtime state files that are not in the data dir (for example comment watermarks): use the documented state path {baseDir}/../state/... until state-location policy changes.

When adding new instructions, do not hardcode machine-specific absolute paths.

Directory Contract

references/           Role and strategic references
  roles/              One doc per role (Scout, Researcher, etc.)
  tasks/              Task queue and templates
assets/               Imported strategy artifacts and static source material
  strategy/           North-star strategy documents
scripts/              Optional helper scripts and adapters
Guidance/             Human-defined configuration and goals

For detailed setup instructions, see Guidance/CONFIGURATION-GUIDE.md.

Cron Job Creation Prompt

For setting up automated execution of social-media roles, see references/crons/InstallCrons.md.

Use one of these paths:

• Basic install: run ./packaged-scripts/install-cron-jobs.sh from this repo root.
• Custom install/tuning: use scripts/install-cron-jobs.sh and references/crons/InstallCrons.md as templates, preserving {baseDir} conventions and role boundaries.