Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Conducts multi-phase academic research using a fixed methodology: two full research cycles per theme, APA 7th edition citations, evidence hierarchy annotations, and three mandatory user checkpoints before the final report. Runs entirely on native OpenClaw tools without external API dependencies.
The mandated two-cycle structure and explicit evidence hierarchy force gap identification and source cross-referencing that single-pass AI research tools skip entirely.
When to use it
- Literature review for a graduate thesis on a contested scientific topic
- Competitive intelligence report on market players before a product launch
- Fact-checking a news article's claims against primary sources
- Synthesizing research on a medical treatment before a patient consultation
- Trend analysis across multiple domains for a strategic planning document
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: What does current research say about muscle mass loss during GLP-1 weight loss treatment?
- 1Ask 2-3 clarifying questions about depth, scope, and source preferences, then wait for user response
- 2Present a structured research plan with 3-5 themes, tool sequence, and expected deliverables, then wait for approval
- 3Run two web_search cycles (count=20) and web_fetch on primary sources for each theme, documenting hypothesis evolution and contradictions between cycles
- 4Synthesize cross-theme patterns, resolve or flag contradictions, and annotate conclusions with confidence levels
- 5Deliver a full narrative report in APA 7th edition with in-text citations and a complete reference list
A multi-section academic report with flowing prose, integrated citations, a contradictions analysis, a limitations section, and an alphabetized APA reference list