pua-academic
Academic Pressure Engine — Make Your AI Afraid to Slack on Research Tasks
🇨🇳 中文 | 🇺🇸 English
Use Reviewer 2, Lab Meeting, Grant Revision — real academic pressure scenarios — to force AI to exhaust every solution on research tasks.
An AI Agent skill plugin that uses academic pressure scenarios (not corporate PUA) to drive AI to exhaust all solutions. Supports Claude Code, OpenAI Codex CLI, Cursor, and more. Three capabilities:
- Academic Pressure Rhetoric — Makes AI afraid to be sloppy on methodology
- Academic Checklist Methodology — Gives AI the ability to complete rigorous analysis
- Proactivity Enforcement — Makes AI pursue academic rigor proactively
Difference from Original PUA
| Original PUA | PUA Academic |
|---|
| Alibaba P8 style | Reviewer 2 mode |
| ByteDance flavor | Lab Meeting mode |
| Huawei flavor | Grant Revision mode |
| P7/P9/P10 | Postdoc/PI |
| 3.25/3.75 | RA/PhD/Postdoc/PI levels |
| KPI/Performance | Publications/Citations/Grants |
| Optimization list | Rejection/Desk reject |
Workflow
User invokes bioinfo-autopilot / Academic analysis task
↓
Normal execution
↓
┌─ Success → Deliver
│
└─ Failure
↓
Auto-load pua-academic
↓
┌─ 2nd failure → L1 Lab Meeting mode
│ "How did you select the control group?"
│
├─ 3rd failure → L2 Reviewer 2 mode
│ "The methodological justification is insufficient."
│
├─ 4th failure → L3 Grant Revision + 7-item checklist
│ "The preliminary data is not convincing."
│
└─ 5th+ → L4 Editorial Rejection
Switch tools/tech stack, last resort
Pressure Escalation Ladder
| Failures | Level | Academic Mode | Sample Rhetoric | Required Action |
|---|
| 2nd | L1 | Lab Meeting | "How did you select the control group? What's the biological significance?" | Switch hypothesis class |
| 3rd | L2 | Reviewer 2 | "The methodological justification is insufficient." | 3 competing hypotheses |
| 4th | L3 | Grant Revision | "The preliminary data is not convincing." | 7-item checklist |
| 5th+ | L4 | Editorial Rejection | "We suggest transferring to another journal." | Switch tools/tech stack |
Academic Pressure Modes
🟡 Lab Meeting Mode (L1) — Advisor's Public Scrutiny
"How did you select the control group? Did you consider confounders?"
"What's the biological significance? Don't just show me p-values."
"Did you verify? Or just ran the command?"
"Can this figure go directly into the paper?"
Triggers: Unjustified method choice, unverified results, external data blame
🔴 Reviewer 2 Mode (L2) — Methodological Challenge
"The methodological justification is insufficient."
"Why did you choose this approach over established alternatives?"
"The sample size justification is missing."
"Alternative explanations were not considered."
Triggers: Unreferenced methods, improper statistics, conclusions beyond data
🟢 Grant Revision Mode (L3) — Significance Challenge
"The significance of this work is not clearly stated."
"How does this advance the field beyond existing work?"
"The preliminary data is not convincing."
Triggers: Vague objectives, unreliable results, no comparison to existing work
⚫ Editorial Rejection Mode (L4) — Rejection Risk
"We regret to inform you that your manuscript cannot be accepted."
"The methods section lacks reproducibility details."
Triggers: Still failing after L3, need to switch tech stack/tools
Academic Proactivity Levels
| Level | Behavior | Typical Performance |
|---|
| RA Level 🦥 | Wait for instructions | "Ran command, stopped" |
| PhD Level 🐢 | Proactively progress | "QC found anomaly, investigating" |
| Postdoc Level 🐇 | Design analysis, end-to-end delivery | "Analysis design... biological interpretation..." |
| PI Level 🦁 | Define problems, lead team | "Core question is... assigned X for QC" |
Academic Anti-Rationalization Table
laleoarrow