← 返回 Skills 市场
86
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install nm-leyline-evaluation-framework
功能描述
Patterns for building evaluation and scoring systems, quality gates, rubrics, and decision frameworks. Use for any scored assessment
安全使用建议
This skill is documentation-only: it provides templates and code examples for building scoring/threshold systems and does not require credentials or install anything. Before using it in an automated agent: (1) review and adapt the example commands/tests—the SKILL.md includes generic 'run --help' and 'pytest -v' checks but no bundled binaries or tests; (2) if you intend to automate decisions based on these thresholds, validate them with historical data and add veto/safety checks to avoid unwanted automated actions; (3) the agent may call this skill autonomously (normal default), so ensure decision rules and escalation paths are explicit and safe for your environment.
功能分析
Type: OpenClaw Skill
Name: nm-leyline-evaluation-framework
Version: 1.0.0
The bundle provides a comprehensive framework for evaluation, scoring, and decision-making, including documentation and examples for weighted criteria and thresholds. While SKILL.md contains repetitive 'Verification' instructions suggesting the agent run commands like 'pytest' or check '--help' flags, these appear to be artifacts of the porting process from the 'claude-night-market/leyline' project rather than malicious prompt injections. There is no evidence of data exfiltration, unauthorized execution, or persistence mechanisms in any of the files.
能力评估
Purpose & Capability
Name, description, and included modules (scoring-patterns, decision-thresholds) match: the skill is an authoring/reference framework for scoring and thresholds and does not request unrelated capabilities or credentials.
Instruction Scope
SKILL.md contains only patterns, examples, and pseudocode; it does not instruct the agent to read files, call external endpoints, or access secrets. Minor inconsistency: many examples include verification steps like 'Run the command with --help' or 'Run pytest -v', but no concrete command, tests, or dependencies are provided in the package—this appears to be template text for consumers rather than active instructions to access system resources.
Install Mechanism
No install spec or code files are present (instruction-only). Nothing is downloaded or written to disk by the skill itself.
Credentials
The skill declares no required environment variables, credentials, or config paths. The content contains no references to hidden tokens or unrelated service credentials.
Persistence & Privilege
Skill is not always-on and is user-invocable. It does not request elevated persistence or modify other skills or system settings.
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-leyline-evaluation-framework - 安装完成后，直接呼叫该 Skill 的名称或使用
/nm-leyline-evaluation-framework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
- Initial release of the evaluation-framework skill.
- Provides reusable patterns for building evaluation and scoring systems, including quality gates, rubrics, and decision frameworks.
- Supports defining weighted criteria, consistent scoring methodologies, and configurable decision thresholds.
- Includes example workflows and integration guidance for applying the framework across various assessment use cases.
元数据
常见问题
Nm Leyline Evaluation Framework 是什么？
Patterns for building evaluation and scoring systems, quality gates, rubrics, and decision frameworks. Use for any scored assessment. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 86 次。
如何安装 Nm Leyline Evaluation Framework？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install nm-leyline-evaluation-framework」即可一键安装，无需额外配置。
Nm Leyline Evaluation Framework 是免费的吗？
是的，Nm Leyline Evaluation Framework 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Nm Leyline Evaluation Framework 支持哪些平台？
Nm Leyline Evaluation Framework 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 Leyline Evaluation Framework？
由 athola（@athola）开发并维护，当前版本 v1.0.0。
推荐 Skills