← 返回 Skills 市场
105
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install nm-attune-war-room-checkpoint
功能描述
Assess decision reversibility and risk at critical checkpoints to determine whether full War Room escalation is warranted
安全使用建议
This skill appears coherent for its purpose, but verify integration details before enabling automated escalation: (1) inspect the night-market.attune:war-room config to see how a 'full War Room' is invoked (endpoint, auth, and what data is transmitted); (2) confirm calling commands implement and respect the skill's requires_user_confirmation flag so actions aren't auto-executed unexpectedly; (3) test in a safe environment with dummy data (note the SKILL.md's `make test-checkpoint` hint assumes repository code not included here); (4) ensure any war-room endpoints or services the config points to are trusted and have least-privilege credentials.
功能分析
Type: OpenClaw Skill
Name: nm-attune-war-room-checkpoint
Version: 1.0.0
The skill bundle defines a structured risk-assessment framework ('War Room Checkpoint') for an AI agent to evaluate decision reversibility. The logic in SKILL.md is purely instructional, focusing on calculating scores and determining escalation paths without any evidence of data exfiltration, malicious execution, or prompt injection. While it references local logging paths and internal Python modules for audit trails, these appear consistent with its stated purpose of maintaining a verifiable decision history.
能力标签
能力评估
Purpose & Capability
Name/description describe checkpointing for escalation. The only declared requirement is a config path (night-market.attune:war-room) which matches the plugin-style integration this skill expects. No unrelated binaries or credentials are requested.
Instruction Scope
The SKILL.md stays largely within the stated purpose: it defines inputs, scoring, thresholds, and the structure of responses for calling commands. Two caveats: (1) it references invoking a 'full War Room' and returning a war_room_session_id but does not specify HOW that external invocation occurs (endpoint, authentication, or data handling) — that behavior is delegated to the calling environment/config, so you should verify the host integration. (2) it suggests running `make test-checkpoint` to verify logic, but this skill bundle contains no code files; that instruction assumes the broader repository context rather than the installed skill. Neither is inherently malicious but both are points to check before trusting automated escalation.
Install Mechanism
Instruction-only skill with no install spec and no code files — nothing is written to disk by the skill itself. This is the lowest-risk install mechanism.
Credentials
The skill requests no environment variables or external credentials. It does require a config path (night-market.attune:war-room) which is appropriate for integration settings; ensure that config does not contain unrelated secrets or tokens you wouldn't want the skill to access.
Persistence & Privilege
always is false and the skill does not request persistent/privileged presence or modify other skills. Autonomous invocation is allowed (platform default) but not excessive here.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install nm-attune-war-room-checkpoint - 安装完成后，直接呼叫该 Skill 的名称或使用
/nm-attune-war-room-checkpoint触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
War Room Checkpoint Skill Initial Release
- Provides inline assessment to evaluate decision reversibility and risk at key command checkpoints.
- Calculates Reversibility Score (RS) and recommends escalation level: express, lightweight, full council, or delphi.
- Returns structured response (with recommendation, rationale, and confidence) for calling commands.
- Supports customizable profile thresholds and command-specific risk adjustments.
- Persists detailed checkpoints and session audit trails for compliance and integrity verification.
- Designed for seamless integration; handles failures gracefully without blocking workflow.
元数据
常见问题
Nm Attune War Room Checkpoint 是什么？
Assess decision reversibility and risk at critical checkpoints to determine whether full War Room escalation is warranted. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 105 次。
如何安装 Nm Attune War Room Checkpoint？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install nm-attune-war-room-checkpoint」即可一键安装，无需额外配置。
Nm Attune War Room Checkpoint 是免费的吗？
是的，Nm Attune War Room Checkpoint 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Nm Attune War Room Checkpoint 支持哪些平台？
Nm Attune War Room Checkpoint 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Nm Attune War Room Checkpoint？
由 athola（@athola）开发并维护，当前版本 v1.0.0。
推荐 Skills