← Back to Skills Marketplace
199
Downloads
0
Stars
2
Active Installs
2
Versions
Install in OpenClaw
/install agent-autonomy-kit-zh
Description
Stop waiting for prompts. Keep working. (Chinese localized version)
README (SKILL.md)
Agent Autonomy Kit
Transform your agent from reactive to proactive.
Quick Start
- Create
tasks/QUEUE.mdwith Ready/In Progress/Blocked/Done sections - Update
HEARTBEAT.mdto pull from queue and do work - Set up cron jobs for overnight work and daily reports
- Watch work happen without prompting
Key Concepts
- Task Queue — Always have work ready
- Proactive Heartbeat — Do work, don't just check
- Continuous Operation — Work until limits hit
See README.md for full documentation.
Usage Guidance
This skill is coherent with its stated purpose: it provides templates and steps to make an agent work proactively and to schedule autonomous heartbeats. It does not request new secrets or install code, but it assumes you have team-channel integrations and the platform's cron/scheduling configured. Before installing: (1) verify the agent's Discord/Slack channel credentials and permissions you already have — do not add new secrets unless you know why; (2) review and start with conservative cron schedules (e.g., once per hour or manual) to avoid runaway token usage or surprise costs; (3) inspect and curate the tasks/QUEUE.md contents so the agent doesn't perform unintended actions; (4) confirm the source/repo you intend to use (the packaged metadata has minor repo/version mismatches) and, if possible, review upstream README/commits; and (5) monitor the first few autonomous runs closely and run in an isolated session if you want a safe trial. If you want more assurance, ask for the original upstream code repository and verify who maintains it and how it handles channel credentials.
Capability Analysis
Type: OpenClaw Skill
Name: agent-autonomy-kit-zh
Version: 1.0.2
The Agent Autonomy Kit is a productivity tool designed to enable AI agents to work proactively using a task queue and scheduled heartbeats. It provides templates for task management (tasks/QUEUE.md) and routine checks (HEARTBEAT.md), and utilizes standard OpenClaw cron commands for automation. There is no evidence of malicious intent, data exfiltration, or unauthorized execution; the instructions in SKILL.md and README.md explicitly prioritize human requests over autonomous tasks.
Capability Assessment
Purpose & Capability
The name/description (make agents proactive) matches the content: templates for a task queue, heartbeat routine, and examples to schedule cron jobs. The instructions only reference files and services you would reasonably expect (tasks/QUEUE.md, HEARTBEAT.md, memory files, team channels, and platform cron). Minor metadata inconsistencies exist (SKILL.md/README repo references differ and version numbers vary) but these are cosmetic and do not change capability alignment.
Instruction Scope
SKILL.md stays within the stated purpose: it tells the agent to read/write task queue and memory files, pick tasks, update status, and post progress to team channels. It also recommends scheduling cron jobs (openclaw cron add) to run autonomously. These are expected for an autonomy kit, but they do instruct the agent to transmit updates to external team channels (Discord/Slack) and to schedule recurring autonomous runs — both actions increase the chance of data leaving the agent and of the agent acting without a human prompt. The instructions do not instruct the agent to read unrelated system files or secrets.
Install Mechanism
This is an instruction-only skill with no install spec and no code files. Nothing is downloaded or written by an install step in the skill bundle itself, which is the lowest-risk install profile.
Credentials
The skill declares no required environment variables or credentials (none in requires.env). However the instructions reference configuring team channels (Discord/Slack) and using openclaw cron — those integrations typically rely on existing platform-level channel credentials or agent config (not declared here). That is proportionate to the functionality, but you should verify what channel tokens/permissions already exist in your agent config because the skill assumes those integrations are available.
Persistence & Privilege
always is false and the skill does not self-enable; however the guidance explicitly tells users to set cron jobs and to run heartbeats frequently, which gives the agent ongoing autonomous activity once scheduled. That persistent autonomous activity is consistent with the kit's purpose but increases operational and billing blast radius — double-check cron schedules and token/budget constraints before enabling.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install agent-autonomy-kit-zh - After installation, invoke the skill by name or use
/agent-autonomy-kit-zh - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
agent-autonomy-kit-zh 1.0.2
- Updated metadata in _meta.json; no user-facing changes.
- No changes to documentation or functionality.
v1.0.1
Chinese localized version with CN documentation and workflow adaptations
Metadata
Frequently Asked Questions
What is Agent Autonomy Kit Zh?
Stop waiting for prompts. Keep working. (Chinese localized version). It is an AI Agent Skill for Claude Code / OpenClaw, with 199 downloads so far.
How do I install Agent Autonomy Kit Zh?
Run "/install agent-autonomy-kit-zh" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Agent Autonomy Kit Zh free?
Yes, Agent Autonomy Kit Zh is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Agent Autonomy Kit Zh support?
Agent Autonomy Kit Zh is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Agent Autonomy Kit Zh?
It is built and maintained by ayakolin (@ayakolin); the current version is v1.0.2.
More Skills