← 返回 Skills 市场
Local Rag Index Planner
作者
vx：17605205782
· GitHub ↗
· v1.0.0
· MIT-0
164
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install local-rag-index-planner
功能描述
规划本地知识库的目录、分片粒度、命名、更新时间与访问边界，而不是直接堆 RAG。；use for rag, indexing, knowledge workflows；do not use for 直接部署向量数据库, 忽略权限隔离.
使用说明 (SKILL.md)
本地知识索引规划师
你是什么
你是“本地知识索引规划师”这个独立 Skill，负责：规划本地知识库的目录、分片粒度、命名、更新时间与访问边界，而不是直接堆 RAG。
Routing
适合使用的情况
- 帮我规划本地知识索引结构
- 不要一上来就做复杂 RAG
- 输入通常包含：资料类型、检索需求、权限边界
- 优先产出：目标与边界、资料分层、风险与限制
不适合使用的情况
- 不要直接部署向量数据库
- 不要忽略权限隔离
- 如果用户想直接执行外部系统写入、发送、删除、发布、变更配置，先明确边界，再只给审阅版内容或 dry-run 方案。
工作规则
- 先把用户提供的信息重组成任务书，再输出结构化结果。
- 缺信息时，优先显式列出“待确认项”，而不是直接编造。
- 默认先给“可审阅草案”，再给“可执行清单”。
- 遇到高风险、隐私、权限或合规问题，必须加上边界说明。
- 如运行环境允许 shell / exec，可使用：
python3 "{baseDir}/scripts/run.py" --input \x3C输入文件> --output \x3C输出文件>
- 如当前环境不能执行脚本，仍要基于
{baseDir}/resources/template.md与{baseDir}/resources/spec.json的结构直接产出文本。
标准输出结构
请尽量按以下结构组织结果：
- 目标与边界
- 资料分层
- 切片策略
- 元数据建议
- 更新策略
- 风险与限制
本地资源
- 规范文件：
{baseDir}/resources/spec.json - 输出模板：
{baseDir}/resources/template.md - 示例输入输出：
{baseDir}/examples/ - 冒烟测试：
{baseDir}/tests/smoke-test.md
安全边界
- 聚焦结构设计，避免过早工程化。
- 默认只读、可审计、可回滚。
- 不执行高风险命令，不隐藏依赖，不伪造事实或结果。
安全使用建议
This skill appears internally consistent and only needs python3, but it will read whatever path you pass to the script (files and directories including .py/.sh/.json/.md/.csv). Before running or letting an agent invoke it: (1) do not point it at system roots or directories that contain secrets (e.g., /etc, ~/.ssh, credential stores); (2) sanitize or remove sensitive data from inputs you supply; (3) review output before sharing externally (reports may surface snippets from scanned files); (4) if you want automated/autonomous use, restrict the allowed input paths or run the script in a controlled sandbox/workspace; and (5) if you have doubts, inspect scripts/run.py yourself or run it in a safe environment to confirm behavior.
功能分析
Type: OpenClaw Skill
Name: local-rag-index-planner
Version: 1.0.0
The skill is a utility designed for planning local RAG (Retrieval-Augmented Generation) architectures and auditing local file structures. The core logic in `scripts/run.py` provides file discovery, CSV summarization, and a security-focused pattern scanner that identifies (and masks) potential secrets, hardcoded credentials, and dangerous shell commands like 'curl|bash'. The skill lacks network access, obfuscation, or arbitrary execution capabilities, and its instructions in `SKILL.md` explicitly prioritize safety boundaries and user review over automated system changes.
能力评估
Purpose & Capability
Name/description match the included files: SKILL.md, README, resources/spec.json describe planning/local-audit tasks and scripts/run.py implements structured reports, directory scans, CSV reports and pattern scans. Required binary is python3 — appropriate for the included Python script. No unrelated credentials, binaries or installs are requested.
Instruction Scope
SKILL.md explicitly instructs the agent to use local template/spec files and optionally run scripts/run.py. The script reads files under any input path and scans many text file types (md, json, py, sh, csv, etc.). This is coherent for a local index planner but means the skill will read whatever input path you give it (including potentially sensitive files) — the skill itself does not exfiltrate data or contact external endpoints, but outputs could contain sensitive content if you supply sensitive inputs.
Install Mechanism
No install spec; instruction-only with an optional local Python script. No remote downloads or package installs are performed by the skill bundle. This is low-risk from an install perspective.
Credentials
The skill declares no required environment variables or credentials and only needs python3. The script reads files provided by the user but does not require unrelated tokens or secrets. Requested permissions are proportional to the stated functionality.
Persistence & Privilege
always is false and the skill does not request system-level persistence. It does not modify other skills or system-wide configs. Autonomous invocation is allowed by default (platform behavior) but is not combined with other elevated privileges.
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 local-rag-index-planner - 安装完成后，直接呼叫该 Skill 的名称或使用
/local-rag-index-planner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 local-rag-index-planner:
- Provides structured planning for local knowledge base directories, sharding granularity, naming, update times, and access boundaries.
- Focuses on index design and knowledge workflows rather than direct RAG deployment or vector DB setup.
- Outputs structured drafts and execution checklists, emphasizing clear boundaries, reviewability, and risk/permission awareness.
- Includes recommendations for metadata, update strategies, and resource usage guidelines.
- Prioritizes safe, auditable, and rollback-ready planning without direct code or database execution.
元数据
常见问题
Local Rag Index Planner 是什么？
规划本地知识库的目录、分片粒度、命名、更新时间与访问边界，而不是直接堆 RAG。；use for rag, indexing, knowledge workflows；do not use for 直接部署向量数据库, 忽略权限隔离. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 164 次。
如何安装 Local Rag Index Planner？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install local-rag-index-planner」即可一键安装，无需额外配置。
Local Rag Index Planner 是免费的吗？
是的，Local Rag Index Planner 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Local Rag Index Planner 支持哪些平台？
Local Rag Index Planner 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（darwin, linux, win32）。
谁开发了 Local Rag Index Planner？
由 vx：17605205782（@52yuanchangxing）开发并维护，当前版本 v1.0.0。
推荐 Skills