360 Secure SkillHub
360 安全技能中心

Academic Deep Research Pro

该 Skill 面向 OpenClaw 宿主,使用其原生网页检索、网页抓取与子会话工具,按每主题两轮研究循环、多来源交叉验证、证据分级与置信度标准执行深度调研,含三次用户确认节点,最终产出带 APA 7 文内引文与参考文献的叙述式研究报告,适用于文献综述、竞品情报等需可复现性的研究任务。

nancliu
nancliu
热度 30.40
最近更新 2026-07-06 16:52:30
分类:数据依赖:内部调用大模型 / 第三方 API应用场景:研究 / 报告撰写 / 市场分析

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add academic-deep-research-pro
Metadata
name
academic-deep-research
slug
academic-deep-research-pro
description
Transparent, rigorous research with full methodology — not a black-box API wrapper. Conducts exhaustive investigation through mandated 2-cycle research per theme, APA 7th citations, evidence hierarchy, and 3 user checkpoints. Self-contained using native OpenClaw tools (web_search, web_fetch, sessions_spawn). Use for literature reviews, competitive intelligence, or any research requiring academic rigor and reproducibility.
homepage
https://github.com/kesslerio/academic-deep-research-clawhub-skill
metadata
{"openclaw":{"emoji":"🔬"}}

Academic Deep Research 🔬

When to Use

Trigger this skill when the user wants:

  • Deep research, exhaustive analysis, or literature review
  • Multi-source verification and evidence hierarchies
  • Academic-style reports with citations

Required Stop Points

  1. Initial engagement: ask 2–3 clarifying questions and confirm understanding.
  2. Research plan: present themes, steps, and deliverables; wait for approval.
  3. Final report: deliver full narrative report with citations.

Minimum Requirements

  • Two full research cycles per theme.
  • Analyze between every tool call.
  • Multiple sources per claim; contradictions addressed.
  • APA 7th in-text citations and reference list.
  • Narrative-only final report (no lists or tables).

References

  • reference/protocol.md — research phases, tool order, and analysis rules
  • reference/writing-style.md — narrative constraints and phase formatting
  • reference/citations-apa.md — citation rules and examples
  • reference/report-template.md — required report structure
  • reference/error-handling.md — gaps, contradictions, failures
  • reference/quality-standards.md — evidence hierarchy and confidence levels
  • reference/parallel-research.md — sessions_spawn workflow
  • quickref.md — short checklist
  • example.md — sample usage