← 返回 Skills 市场
Shadows Deep Research
作者
NakedoShadow
· GitHub ↗
· v1.1.0
333
总下载
1
收藏
2
当前安装
2
版本数
在 OpenClaw 中安装
/install shadows-deep-research
功能描述
4-round deep research protocol — Explorer, Devil's Advocate, Multi-Perspective Synthesis, Crystallize. Use for major architectural decisions, technology choi...
安全使用建议
This skill is instruction-only and coherent with its description: it asks the agent to run a 4-round research process and requires no installs, secrets, or filesystem access. Before using it, confirm how your platform's 'built-in search' works (queries may go to external search services) and avoid submitting proprietary secrets or sensitive data as part of prompts. If you care about provenance, verify the cited sources the agent provides and consider running any recommended PoCs in an isolated environment. Lastly, note the skill's author/homepage metadata is external — if you want greater assurance, check that external link and the publisher's reputation.
功能分析
Type: OpenClaw Skill
Name: shadows-deep-research
Version: 1.1.0
The skill bundle is a pure reasoning protocol designed to guide an AI agent through a structured four-round research process (Explorer, Devil's Advocate, Synthesis, and Crystallize). It contains no executable code, scripts, or external dependencies, and explicitly states it does not require file system or network access beyond the agent's standard search tools. The instructions in SKILL.md are focused on analytical methodology and do not contain any malicious prompt injection or data exfiltration attempts.
能力评估
Purpose & Capability
The name/description (deep research for architecture/tech decisions) matches the SKILL.md protocol and required resources: it's purely procedural and requests no binaries, env vars, or config paths.
Instruction Scope
The runtime instructions stay within the stated purpose (four rounds of research, synthesis, and citation). The SKILL.md claims 'zero network access' but also says it will be 'evidence-based' and use the agent's built-in search tools — a minor ambiguity about whether the agent will perform remote searches. This is not a security violation, but you should understand that fulfilling the 'evidence-based' rule depends on the agent platform's search capability.
Install Mechanism
No install spec and no code files — instruction-only skill with no on-disk installation risk.
Credentials
No environment variables, credentials, or config paths are requested; requested access is proportionate to the stated function.
Persistence & Privilege
always is false, the skill is user-invocable and does not request permanent presence or system configuration changes.
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 shadows-deep-research - 安装完成后，直接呼叫该 Skill 的名称或使用
/shadows-deep-research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
v1.1.0
HIGH TRUST: PREREQUISITES + SECURITY CONSIDERATIONS + homepage
v1.0.0
Initial release: Implements a structured 4-round protocol for deep research and major technical decisions.
- Outlines when to trigger or skip the protocol.
- Defines four rounds: Explorer, Devil’s Advocate, Multi-Perspective Synthesis, and Crystallize.
- Establishes strict rules: evidence-based analysis, contrarian thinking, and no early narrowing.
- Provides a clear, actionable output format for results and recommendations.
元数据
常见问题
Shadows Deep Research 是什么？
4-round deep research protocol — Explorer, Devil's Advocate, Multi-Perspective Synthesis, Crystallize. Use for major architectural decisions, technology choi... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 333 次。
如何安装 Shadows Deep Research？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install shadows-deep-research」即可一键安装，无需额外配置。
Shadows Deep Research 是免费的吗？
是的，Shadows Deep Research 完全免费（开源免费），可自由下载、安装和使用。
Shadows Deep Research 支持哪些平台？
Shadows Deep Research 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 Shadows Deep Research？
由 NakedoShadow（@nakedoshadow）开发并维护，当前版本 v1.1.0。
推荐 Skills