clawdo
Todo list and task management for AI agents. Add, track, and complete tasks with autonomy levels — agents propose work, humans approve. Works in heartbeats, cron, and conversations. Persistent SQLite CLI with structured JSON output.
Why use this skill?
Clawdo is a persistent task management tool for AI agents. Featuring autonomy levels, SQLite storage, and JSON output, it helps agents track and execute work.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/lepetitpince/clawdoWhat This Skill Does
clawdo is a specialized persistent task management system designed specifically for the autonomous agent lifecycle. Unlike human-centric todo lists, clawdo bridges the gap between agent memory, cron processes, and real-time conversation. It maintains a persistent state via an internal SQLite database, providing structured JSON outputs that allow agents to reason about, prioritize, and execute tasks without human-like terminal interpretation.
The core strength of clawdo lies in its autonomy hierarchy. By categorizing tasks into 'auto', 'auto-notify', and 'collab' tiers, it allows developers to define the exact level of risk an agent is permitted to take. It enforces a strict safety-first protocol: agents propose tasks, humans approve them, and autonomy levels can only be demoted (never promoted) by the system in response to task failures, ensuring that as an agent proves unreliable, it is automatically throttled into a supervised mode.
Installation
Installation is streamlined for the OpenClaw ecosystem:
- Open your workspace and run
clawhub install openclaw/skills/skills/lepetitpince/clawdoto bundle the skill and its documentation. - Ensure Node.js ≥18 is present in your environment.
- Globally install the CLI:
npm install -g clawdo.
Use Cases
- Autonomous Maintenance: Use clawdo within heartbeat loops to ensure agents pick up low-risk background maintenance tasks (like linting or dependency updates) when the agent is otherwise idle.
- Deferred Workflows: Capture requests during chaotic conversations (e.g., 'J mentioned fixing the auth module') and store them as proposed tasks for later triage.
- Inter-Agent Task Delegation: Sub-agents can use clawdo to pass tasks to the main loop, allowing for complex asynchronous workflows that survive agent restarts.
- Human-in-the-Loop Supervision: Use the proposed state for high-stakes tasks where you want to review the agent's plan before execution begins.
Example Prompts
- "I've just added three new tasks to the queue. List all proposed items and tell me which ones you think we should promote to auto-notify status."
- "Check the current inbox. If there are any urgent tasks related to the production database, mark them for my review and flag them as collab."
- "We have a downtime window coming up. Go through the clawdo list, execute all pending auto-notify tasks that don't require network access, and give me a summary of your actions."
Tips & Limitations
- Always parse JSON: When integrating into custom agent scripts, always use the
--jsonflag to avoid parsing terminal formatting which can lead to brittle automation. - Safety First: Don't be afraid to set tasks as 'collab'. The 'auto' tiers are for high-confidence, low-risk actions. If an agent hits three failures, the system will downgrade the task automatically—don't try to override this behavior, as it is designed to protect your environment.
- Contextual Awareness: Remember that clawdo is persistent. If your agent is destroyed or restarted, the queue remains intact, making it ideal for long-running infrastructure maintenance tasks.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-lepetitpince-clawdo": {
"enabled": true,
"auto_update": true
}
}
}Tags
Flags: file-write, file-read
Related Skills
markdown-knowledge
将本地 Markdown 知识库与 OpenClaw 集成,支持语义检索和上下文注入。仅在用户触发时检索(搜索知识库、查一下知识库等),不主动注入。
calling-agent-squad
Activate a multi-agent team (the Squad) to manage complex projects, business tasks, or development workflows. The squad includes a Manager, Architect, Coder, Reviewer, and Observer. Use when the user wants to "call a squad", "start a project", or "deploy squad" with specialized roles and quality control loops.
Claude Code CLI for OpenClaw
Install, authenticate, and use Claude Code CLI as a native coding tool for any OpenClaw agent system.
context-compressor
Intelligently compress context — conversations, code, logs. Preserve key information while reducing token usage. Auto-detects content type and applies optimal compression.
daily-report-generator
Automatically generate daily/weekly work reports from git commits, calendar events, and task lists. Use when you need to quickly create professional work reports without manual effort.