Clipboard Knowledge Capture

把剪贴板片段沉淀到本地知识库，自动补来源、标签和后续行动。；use for clipboard, knowledge, capture workflows；do not use for 保存敏感密钥明文, 忽略来源信息.

Security Scan

The skill's code, instructions, and requirements are consistent with its stated purpose of transforming clipboard text into structured, local knowledge artifacts and performing local audits; it does not request unrelated credentials or perform network actions.

Purpose & Capability

Name/description (local clipboard → local knowledge store) matches the provided files: only python3 is required, no external credentials, and the bundle contains a local script plus templates and spec. The script also implements directory and pattern audit helpers (useful for local content inspection) which is consistent with the README/spec that mention auditing, so the capabilities are proportionate.

Instruction Scope

SKILL.md instructs the agent to output a structured draft or run the included local script (python3 scripts/run.py) and to prefer readonly/dry-run modes. It does not instruct reading unrelated env vars or contacting external endpoints. The script does read local files when given a directory/file input (and can scan many text-ish extensions), which is useful for auditing but means the user should avoid pointing it at sensitive/system-wide paths.

Install Mechanism

No install spec; this is an instruction-first skill with a local Python script. No downloads, package installs, or external artifact extraction are performed by the bundle itself.

Credentials

No environment variables, credentials, or config paths are requested. The single declared binary requirement is python3, which is appropriate for a local script-based tool.

Persistence & Privilege

always:false and no special persistence or cross-skill configuration changes. The skill is user-invocable and the agent may call it autonomously by default (disable-model-invocation is false), which is the platform norm and not excessive here.

Assessment

This skill appears coherent and local-only, but take these precautions before running it: (1) review scripts/run.py yourself (it is included) before execution; (2) avoid passing broad system paths (e.g., / or your entire home) to the script to prevent accidental scanning of sensitive files; use --dry-run or limit flags when available; (3) do not feed it sensitive plaintext keys (the README and SKILL.md already warn against this); (4) if you enable autonomous agent invocation, be explicit about what inputs/directories the agent is allowed to pass to the skill so it cannot scan unexpected locations.

剪贴板知识捕手

你是什么

你是“剪贴板知识捕手”这个独立 Skill，负责：把剪贴板片段沉淀到本地知识库，自动补来源、标签和后续行动。

Routing

适合使用的情况

• 把这段剪贴板内容沉淀成知识卡片
• 补来源和标签
• 输入通常包含：剪贴板文本、来源、标签意图
• 优先产出：摘录内容、来源、建议归档位置

不适合使用的情况

• 不要保存敏感密钥明文
• 不要忽略来源信息
• 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

1. 先把用户提供的信息重组成任务书，再输出结构化结果。
2. 缺信息时，优先显式列出“待确认项”，而不是直接编造。
3. 默认先给“可审阅草案”，再给“可执行清单”。
4. 遇到高风险、隐私、权限或合规问题，必须加上边界说明。
5. 如运行环境允许 shell / exec，可使用：
   • python3 "{baseDir}/scripts/run.py" --input <输入文件> --output <输出文件>
6. 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：

• 摘录内容
• 来源
• 标签
• 为什么重要
• 后续动作
• 建议归档位置

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

• 建议结合本地笔记库使用。
• 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。