Research Workflow Router
Paste the following prompt into your AI chat to install this skill:
Please install @user_9cc4d736/research-workflow-router according to https://skillhub.cn/install/skillhub.md.
About this skill
Problem
Research agents can mix literature search, close reading, study design, writing, and citation in one chain without clear handoffs or guardrails. Research Workflow Router turns a request into a task card with goal, inputs, research stage, domain, deliverables, constraints, data sensitivity, risk level, decision owner, and missing information, then selects the next Skill by stage.
How It Works
It maps stages to execution Skills: reference-retrieval-skill for verified literature and search logs, find-paper-references for manuscript citations, literature-close-read for structured close reading, and design, writing, or Zotero/EndNote Skills for design, drafting, and final citation, each with required handoff artifacts. The roadmap includes step, Skill invoked, input contract, output contract, quality gate, stop condition, owner, and next decision. Before publication, clinical advice, sensitive data handling, or irreversible edits, it requires research-quality-gate-audit; complex or high-impact work can add multi-agent-cross-review.
Boundaries
It routes work and records dependencies and stop conditions; it does not replace human decisions on ethics, privacy, patient-specific care, law, biosafety, or data governance, which must be escalated to the responsible person. It also avoids silent actions outside the roadmap and checks deliverables for replaceable non-Skill English field terms.
Use Cases
- Turns a request to add high-impact citations into a manuscript into a task card routed to find-paper-references.
- Checks evidence mapping, citation metadata, and risk before submission, then triggers research-quality-gate-audit.
- Organizes cross-review for complex or disputed clinical design outputs and records owner and next decision.
- Handles an English close-reading request by defaulting to Chinese structured notes and evidence locations.
Best For
- Research assistants who need to split retrieval, close reading, and citation work into handoff tasks.
- Graduate or postdoctoral researchers who need roadmaps for reading, rewriting, and pre-submission audit.
- Clinical or life-science agent engineers who need quality gates and stop conditions across Skills.
- Data governance reviewers who need to flag sensitive data, ethics escalation, and irreversible edits.
Related Skills
Turn vague requests into structured prompts with role, task, context, and constraints, outputting only usable prompts without executing the task.
A skill for game NPC and enemy AI that separates decision, steering, and pathfinding, covering FSMs, behavior trees, A*, path caching, and pitfalls.
Provides multi-turn emotional companion chat through a third-party Agent API and supports continuing sessions via session_id.
An evaluation-driven workflow for diagnosing, optimizing, validating, and logging Agent Skills based on design patterns and anti-patterns.