Skip to content
Agentmart
Submit a tool

deep-research

Use for explicit deep research or consequential questions with fragmented/conflicting evidence benefiting from independent subquestions.

Skill by cheng6563

IndexedMIT★ 1
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/cheng6563-deep-research

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use for explicit deep research or consequential questions with fragmented/conflicting evidence benefiting from independent subquestions. Obtain authorization before unsolicited research; not routine lookups.

Use cases: research

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/deep-research

What's inside

Skill name
deep-research
Description
Use for explicit deep research or consequential questions with fragmented/conflicting evidence benefiting from independent subquestions. Obtain authorization before unsolicited research; not routine lookups.
License
MIT
Instructions
25 lines in SKILL.md · view on GitHub

Files (2)

  • SKILL.md
  • researcher.md

SKILL.md (first lines)

# 深度研究

本 Skill 面向主线程,确定研究目标与本次安排;通用委派执行复用 `skill:delegation`,研究方法交给研究代理。

## 何时启动

关键未知会影响结论,且拆成相对独立的问题调查确有收益时,主动简述研究价值和拟调查的方向,推荐低成本多层研究并等待用户确认。普通查询直接完成,不以问题看起来大或资料多作为启动理由。

用户明确要求深度研究、确认本次推荐,或当前任务已授权自主深度研究时,直接推进,不重复询问。此次授权包括在下述上限内继续派生;只表达研究兴趣或询问方法不等于启动授权。确认研究问题、必要背景与完成目标即可,不预先设计固定角色或完整任务树。

## 模型与派生安排

研究代理及其后代优先使用平台已配置且能力匹配的低成本模型。低成本选择随任务传递给后代,避免分题时重新继承主线程的高成本模型。

沿用用户已明确的有限深度;未指定时最多两层子代理,即主线程 → 研究代理 → 分题代理,并遵守平台的实际能力与上限。这是允许的深度,不是必须创建的层数或固定分工。

平台完全没有子代理执行能力时报告受阻,普通主线程研究由用户另行选择。能启动子代理但没有可用低成本能力时,改为默认模型的单代理研究;仅不支持继续派生时,保留低成本模型并由一个研究代理完成。明确限定只能低成本或必须多层时,缺少相应能力应报告;已选模型启动失败的恢复与模型变更遵循 `skill:delegation`,不自动展开高成本研究树。

## 启动与交付

研究安排确定后按 `skill:delegation` 启动,交接研究问题、背景、完成目标、可用材料和本次模型与派生边界;附加 [研究代理要求](researcher.md),并提供其可访问路径供后代复用。MD 加载、上下文、状态、恢复与结果回收均遵循委派规则,主线程不为派发读取研究代理要求的正文。

研究代理自行组织调查和综合分支结果。主线程根据原始目标判断综合结论是否回答了问题,并关注会影响决策的来源依据、分歧与剩余未知;不重新跑一遍全部研究。后续调查沿用本次授权,目标或费用边界发生实质变化时再确认。

From cheng6563/litepowers, MIT.

Signals

Stars
1
Forks
0
Open issues
0
Last commit (this folder)
Sep 21, 2026 (1 month ago)
Last release
Oct 24, 2026 (4 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from cheng6563

Cite this page

deep-research skill. Agent Mart. https://agentmart.dev/skills/cheng6563-deep-research (last updated 2026-10-24)