TenureAIOfficial·9 Agent Skills Included

PhD-Zero

Structured research planning, experiments, memory, and paper writing

Runs end-to-end research workflows: literature surveys, experiment planning, training execution, and paper drafting. Eliminates improvised answers, lost context, and abandoned long-running jobs through enforced routing and durable memory. Keeps every step evidence-based with stage reports, visualizations, and human checkpoints for risky decisions.

51|4

↗

npx skills add TenureAI/PhD-Zero --all -g -y

Available:

Instructs your AI agent to route every research task through the correct sub-skill, enforce evidence-based execution, monitor long-running jobs, and escalate risky decisions to a human.

All Skills in This Repository (9)

Pure Emerald Level Indicators

📦 In Repo

TenureAI

human-checkpoint

Escalate critical AI R&D decisions to human operators for approval.

research-workflow

Orchestrate mode-aware, evidence-driven AI R&D workflows from intake to completion.

run-governor

Manage AI research run execution policies, interaction modes, and safety allowances.

deep-research

Conduct auditable deep research with staged searches, evidence verification, and traceable citations.

research-plan

Generate execution-ready research plans with experiments and expected outcomes.

memory-manager

Retrieve and write back AI R&D memory across working, episode, procedure, insight, and persona types.

experiment-execution

Execute AI/ML experiments locally or remotely with explicit environment and runtime controls.

project-context

Initialize and maintain per-project private runtime context for AI R&D tasks.

paper-writing

Draft and revise academic paper sections with evidence-first writing and LaTeX output.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install PhD-Zero?▼

Run `npx skills add TenureAI/PhD-Zero --all -g -y` in your terminal to install all research skills globally.

What does PhD-Zero do?▼

It gives your coding agent a disciplined research workflow covering literature review, experiment planning, execution monitoring, memory, and paper writing.

Can PhD-Zero run long training jobs autonomously?▼

Yes. It tracks long-running jobs with durable action records, polls their status, diagnoses failures, and keeps working until targets are met.

Does PhD-Zero work with Claude Code and Codex?▼

Yes. The same skill library is exposed through AGENTS.md for Codex-style agents and mirrored under .claude/skills for Claude Code.

Can PhD-Zero help write a research paper?▼

Yes. Its paper-writing skill drafts and revises abstracts, introductions, related work, methods, and experiment sections from your actual evidence and citations.