Security audit
Security checks for vulnerabilities and agentic risk
This skill is an agent-monitoring and coordination helper whose command use is disclosed and aligned with its purpose.
Before installing, understand that this skill can display local OpenClaw agent, workspace, status, and session information, and its communication script can send a message to an agent you name. Use it only with agents and workspaces you control, and treat --watch as a foreground monitoring loop you should stop when no longer needed.
"""发送消息给指定智能体"""
try:
# 使用 sessions_send 发送消息
result = subprocess.run(
['openclaw', 'sessions', 'send',
'--agent', to_agent,
'--message', message],def get_agent_status():
"""获取智能体状态"""
try:
result = subprocess.run(
['openclaw', 'status', '--json'],
capture_output=True,
text=True,def get_session_history(session_key):
"""获取会话历史"""
try:
result = subprocess.run(
['openclaw', 'sessions', 'history', session_key, '--json'],
capture_output=True,
text=True,def get_agents():
"""获取所有智能体列表"""
try:
result = subprocess.run(
['openclaw', 'agents', 'list', '--json'],
capture_output=True,
text=True,def get_agent_sessions(agent_id):
"""获取智能体的会话"""
try:
result = subprocess.run(
['openclaw', 'sessions', 'list', '--json'],
capture_output=True,
text=True,def get_agent_sessions(agent_id):
"""获取智能体的会话"""
try:
result = subprocess.run(
['openclaw', 'sessions', 'list', '--json'],
capture_output=True,
text=True,No suspicious patterns detected.