Classify research requests into proceed or refuse-with-scaffold behaviors.
npx skills add https://github.com/alompson/phd-guardrail-skill --skill phd-guardrail
Please help me install this Agent Skill. Skill: phd-guardrail Source: https://github.com/alompson/phd-guardrail-skill/tree/main Command: npx skills add https://github.com/alompson/phd-guardrail-skill --skill phd-guardrail
It prevents PhD researchers from outsourcing core intellectual work—like literature synthesis, gap-finding, and thesis writing—to a chatbot, which can cause cognitive debt and atrophy of researcher skills.
Use this skill whenever you ask an AI to help with your thesis, research question, literature review, gap analysis, or novelty assessment so it can decide what to delegate versus scaffold.
High-intent search queries and answers about installing and using this skill.
To prevent AI from doing your PhD literature synthesis, a guardrail classifies requests as RED, AMBER, or GREEN, refusing to draft core academic thinking and instead providing Socratic scaffolding and structured reading protocols so you complete the synthesis yourself.
Socratic scaffolding for academic writing is a guided protocol that refuses to generate your thesis text directly, instead providing structured prompts and reading frameworks that help you independently develop research questions, gap analyses, and novelty assessments.
You can use AI for thesis formatting and retrieval because the guardrail triage classifies these as safe GREEN tasks, while actively refusing and scaffolding RED tasks like gap-finding and literature synthesis to prevent cognitive debt and skill atrophy.
Research question requests are evaluated through operational classification routing them into proceed or refuse-with-scaffold behaviors, ensuring safe retrieval proceeds while core PhD thinking tasks like novelty assessments are scaffolded to protect your intellectual development.
The guardrail limits AI from drafting literature reviews, gap analyses, and thesis writing to prevent partial-synthesis risks, meaning you must independently read and synthesize materials rather than adopting AI-constructed understanding as your own academic work.
The PhD guardrail requires YAML-defined discovery metadata plus operational classification to function, allowing it to accurately route your academic writing and research requests into proceed versus refuse-with-scaffold behaviors based on task type.
tachyon-beepRoute Python problems to the appropriate specialist skill by symptom classification.
catlog22Formulate research topics into structured plans with literature gap analysis.
roycolumbia-codeGenerate AI-SPEC.md design contracts and orchestrate workflow tasks.
rakibulismDevelop scholarly research plans and literature syntheses from a topic.
