Stop waiting for prompts. Keep working. - Automaton Edition

This skill appears to do what it says: turn an agent proactive by reading local task files, running periodic 'heartbeat' workflows, posting progress to team channels, and scheduling cron jobs. Before installing: 1) Decide whether you want an agent to run tasks automatically and how often — frequent heartbeats can consume tokens and perform actions without immediate human review. 2) Verify and provision any external credentials (Discord/Slack tokens, openclaw cron permissions) in a controlled place (not pasted into the skill) and restrict what the agent may post. 3) Test in an isolated or low-privilege agent/session first (use an 'isolated' session as the README suggests). 4) Add rate limits, token-budget safeguards, and a clear human-override mechanism so the agent stops or reports before performing high-impact actions. 5) Audit any cron jobs you add and review what the agent writes to memory/task files to avoid accidental leakage of secrets. If you need more assurance, ask the author for an explicit list of required runtime credentials and a sample configuration that shows where those credentials are stored and how human approval is enforced.

Type: OpenClaw Skill Name: agent-autonomy-kit-automaton Version: 1.0.2 The skill bundle is a productivity framework designed to enable agent autonomy through a task queue system and proactive heartbeat routines. It provides markdown templates (QUEUE.md, HEARTBEAT.md) and instructions for setting up cron jobs using the 'openclaw' CLI to allow the agent to perform tasks without direct human intervention. The instructions are transparent, aligned with the stated purpose of maximizing token utilization, and contain no evidence of data exfiltration, malicious execution, or harmful prompt injections.