← All concepts

agent control patterns

3 articles · 4 co-occurring · 0 contradictions · 5 briefs

Agent 尝试访问白名单外的主机时，OpenShell 拦截请求并在 TUI 中弹出审批，实现"人在回路"的安全模式" — Demonstrates human-in-the-loop pattern where system intercepts unauthorized access attempts and prompts user for approval

Related concepts
system prompt architecture 1 resource isolation 1 prompt engineering 1 agent team alignment 1
Signal history
2026-W15
15
Evidence chain (3 articles, showing 3)
@shao__meng: OpenClaw 正在成为个人 AI 的"操作系统"，一个始终在线、能自我进化的自主助手。但自主性越强，安全风险越大：Agent 可能访问不该访问的网络、... example_of

Agent 尝试访问白名单外的主机时，OpenShell 拦截请求并在 TUI 中弹出审批，实现"人在回路"的安全模式" — Demonstrates human-in-the-loop pattern where system intercepts unauthorized access attempts and prompts user for approval

@doodlestein: The clankers need to understand who is running things. Otherwise they get ann... example_of

[INFERRED] "this is why I have my RULE 0 in all my AGENTS .md files" — Author demonstrates a practical control mechanism (RULE 0) implemented in agent configuration files to override default LLM const

@scottbelsky: thinking: products that help humans get credit for the work accomplished by a... supports

[INFERRED] "products that help humans get credit for the work accomplished by agents they supervise in the enterprise will have better adoption" — Article argues that agent-assisted enterprise solutio

query this concept
$ db.articles("agent-control-patterns")
$ db.cooccurrence("agent-control-patterns")
$ db.contradictions("agent-control-patterns")