Log agent activities to Mission Claw dashboard with token usage tracking. Use when completing significant tasks, subagent work, or any action worth recording in the activity feed. Triggers on task completion, activity logging, or tracking agent work.
数据来源:ClawHub。 在 ClawSkills 查看
选择你使用的 Agent
方法一:命令行安装(推荐)
推荐(无需提前安装 clawhub)
npx clawhub@latest --dir ~/.claude/skills install mission-claw或使用 clawhub CLI(需提前安装)
clawhub --dir ~/.claude/skills install mission-claw⚠️ 需要 Node.js 18+,没有 Node?请使用下方方法二直接下载 ZIP。 安装 Node.js →
方法二:手动下载安装(无需 Node)
下载 ZIP,解压后将文件夹放到以下路径,重启 Agent 即可:
安装路径
~/.claude/skills/mission-claw/💡解压后将文件夹放到上方路径,重启 Agent 即可生效
--- name: mission-claw description: Log agent activities to Mission Claw dashboard with token usage tracking. Use when completing significant tasks, subagent work, or any action worth recording in the activity feed. Triggers on task completion, activity logging, or tracking agent work. binaries: [mclaw] install: npm install -g mission-claw homepage: https://github.com/tsangwailam/mcclaw repository: https://github.com/tsangwailam/mcclaw ---
Log significant tasks to the Mission Claw activity feed for tracking agent work, token usage, and project progress.
mclaw binary must be installed on the host system.session_status tool to retrieve current token usage.mclaw log "Task description" \
--agent "AgentName" \
--project "ProjectName" \
--status completed \
--duration "Xm Ys" \
--input-tokens N \
--output-tokens N \
--total-tokens N
action - Brief task description (positional argument)--agent - Agent name (e.g., "J", "mission-claw", "type-alchemy")--total-tokens - Total tokens used (always include)--details - Longer description of work done--project - Project name--status - completed (default), in_progress, or failed--duration - Time taken (e.g., "5m", "1h 30m")--input-tokens / --output-tokens - Token breakdownLog after completing:
Before logging, check your session's token usage:
session_status tool to see current tokens# Feature implementation
mclaw log "Added date-time filter to dashboard" \
--agent "mission-claw" \
--project "Mission Claw" \
--status completed \
--duration "10m" \
--total-tokens 15000
# Quick fix
mclaw log "Fixed timezone bug" \
--agent "J" \
--project "ContentMorph" \
--duration "2m" \
--total-tokens 3500
# In-progress work
mclaw log "Implementing payment flow" \
--agent "J" \
--project "TypeAlchemy" \
--status in_progress
mclaw list # Recent activities
mclaw list --agent J # Filter by agent
mclaw status # Quick stats and service status
mclaw dashboard start # Start web dashboard (localhost:3101 by default)
The CLI logs to a local daemon (defaulting to http://localhost:3100). You can also POST directly to the API:
POST http://localhost:3100/api/activity
{
"action": "Task name",
"agent": "J",
"project": "Project",
"status": "completed",
"totalTokens": 5000,
"inputTokens": 4000,
"outputTokens": 1000
}安装 Mission Claw 后,可以对 AI 说这些话来触发它
Help me get started with Mission Claw
Explains what Mission Claw does, walks through the setup, and runs a quick demo based on your current project
Use Mission Claw to log agent activities to Mission Claw dashboard with token usage tra...
Invokes Mission Claw with the right parameters and returns the result directly in the conversation
What can I do with Mission Claw in my data & analytics workflow?
Lists the top use cases for Mission Claw, with example commands for each scenario
将技能文件夹放到 ~/.claude/skills/mission-claw/ 目录(个人级,所有项目可用),或 .claude/skills/mission-claw/(项目级)。重启 AI 客户端后,用 /mission-claw 主动调用,或让 AI 根据上下文自动发现并使用。
Mission Claw 支持 Claude、Cursor、OpenClaw,可与这些 AI 平台无缝集成,扩展其能力。
Mission Claw 可免费安装使用。请查阅仓库了解许可证信息。
Log agent activities to Mission Claw dashboard with token usage tracking. Use when completing significant tasks, subagent work, or any action worth recording in the activity feed. Triggers on task completion, activity logging, or tracking agent work.
Mission Claw 属于「Data & Analytics」分类,该分类的技能帮助 AI 智能体在此领域执行专业任务。
Automate my data & analytics tasks using Mission Claw
Identifies repetitive steps in your workflow and sets up Mission Claw to handle them automatically