Agent Skills

academic-research-skills-codex

AI-assisted academic research workflows for literature review, paper writing, peer review, and research pipelines

Install

npx skills add https://github.com/reason-machines/codex-skills --skill academic-research-skills-codex
SKILL.md

Academic Research Skills for Codex

Skill by ara.so — Codex Skills collection.

This skill provides a comprehensive suite of AI-assisted academic research workflows including deep research, paper writing, peer review, and end-to-end research pipelines. Originally designed for Claude Code, this Codex-native package routes requests to specialized research agents based on user intent.

What It Does

Academic Research Skills (ARS) for Codex provides five core workflows:

  1. Deep Research - Literature reviews, systematic reviews, research question refinement, fact-checking
  2. Academic Paper - Outlining, drafting, citation formatting, revision, AI disclosure
  3. Academic Paper Reviewer - Manuscript review, simulated peer review, editorial decisions
  4. Academic Pipeline - End-to-end research-to-publication workflow with integrity gates
  5. Experiment Agent - Code experiment planning, study protocols, reproducibility validation

Installation

Install using the Codex skill installer with git method:

python "$HOME/.codex/skills/.system/skill-installer/scripts/install-skill-from-github.py" \
  --repo Imbad0202/academic-research-skills-codex \
  --ref main \
  --path skills/academic-research-suite \
  --method git

Update Existing Installation

rm -rf "$HOME/.codex/skills/academic-research-suite"
python "$HOME/.codex/skills/.system/skill-installer/scripts/install-skill-from-github.py" \
  --repo Imbad0202/academic-research-skills-codex \
  --ref main \
  --path skills/academic-research-suite \
  --method git

After installation, open a new Codex conversation to activate the skill.

Verify Installation

/skills

You should see one entry: academic-research-suite. You should NOT see separate entries for academic-paper, deep-research, etc.

Core Usage Patterns

Basic Invocation

Use $academic-research-suite followed by your research task:

Use $academic-research-suite to help me plan a systematic literature review on
AI adoption in higher education quality assurance.

The Codex adapter automatically routes to the appropriate workflow based on your request.

Workflow Selection

Workflow When to Use Example Trigger
deep-research Need research question refinement, literature review, fact-checking "conduct a systematic review on..."
academic-paper Need paper outline, drafting, citations, revision "write a research paper about..."
academic-paper-reviewer Need manuscript review, peer review simulation "review this research manuscript..."
academic-pipeline Need full research-to-publication workflow "run end-to-end pipeline from topic to paper"
experiment-agent Need code experiment design, study protocols "plan a reproducible code experiment..."

Structured Request Pattern

For best results, provide goal, current state, and constraints:

Use $academic-research-suite.

Goal: write a journal article on AI in higher education QA
Current materials: literature matrix and rough findings, no outline yet
Output needed: paper architecture and evidence checklist
Constraints: English, APA 7, policy audience

Command Aliases

Codex emulates Claude Code /ars-* commands as aliases:

Alias Workflow Purpose
ars-plan academic-paper Create paper plan/structure
ars-outline academic-paper Generate outline only
ars-abstract academic-paper Draft abstract
ars-lit-review academic-paper Build literature review
ars-citation-check academic-paper Verify citations
ars-disclosure academic-paper Generate AI disclosure statement
ars-format-convert academic-paper Convert between formats
ars-revision-coach academic-paper Get revision guidance
ars-revision academic-paper Execute revisions
ars-full academic-pipeline Run full pipeline

Using Aliases

Either prefix style works:

Use $academic-research-suite: ars-plan my paper on AI governance

Or if your client supports it:

/ars-plan my paper on AI governance

Workflow Examples

Deep Research: Socratic Question Refinement

When you have a broad topic but no clear research question:

Use $academic-research-suite.

I want to write about AI adoption in higher education quality assurance.
I don't have a clear research question yet.
Use Socratic dialogue to help me narrow the question first.

Expected: ARS asks narrowing questions before producing an outline.

Academic Paper: From Notes to Draft

Use $academic-research-suite.

Task: Turn these research notes into an IMRaD paper.
Materials: [attach notes.md, data_summary.csv]
Target: 6000-8000 words, APA 7, submit to Higher Education Research
Output: Full outline with section word counts, then draft intro

Paper Review: Simulated Peer Review

Use $academic-research-suite to review this manuscript.

Mode: full review
Focus: methodology, contribution, citation integrity, desk-reject risks
Output: reviewer reports + editorial decision letter

[attach manuscript.pdf or manuscript.md]

Academic Pipeline: Staged Execution

For checkpoint-based workflow:

Use $academic-research-suite to start an academic-pipeline run.

Begin with Stage 0 intake and stop after producing the pipeline dashboard.
Wait for my approval before Stage 1.

Experiment Agent: Reproducible Study Design

Use $academic-research-suite experiment-agent.

Plan a code experiment comparing three ML models on education dataset.
Requirements: reproducible, statistical rigor, version control, artifact storage
Output: experiment protocol + reproducibility checklist

Configuration

Environment Variables

For cross-model verification (optional):

export ARS_CROSS_MODEL=claude-opus-4.7
export ANTHROPIC_API_KEY=your_api_key_here

Without these, ARS runs entirely within Codex/OpenAI. Cross-model review uses Anthropic API when explicitly requested.

Material Passport

ARS tracks research artifacts through "Material Passports" - metadata files that follow:

{
  "artifact_id": "paper_v1_outline",
  "created": "2026-05-17T10:30:00Z",
  "workflow": "academic-paper",
  "stage": "outline",
  "integrity_checks": ["citation_format", "structure"],
  "sources": ["source1.pdf", "source2.pdf"],
  "status": "approved"
}

Reset Material Passport by starting a new Codex conversation.

Working with Citations

Citation Check

Use $academic-research-suite: ars-citation-check

[paste paper text with citations]

Verify all citations are formatted correctly in APA 7.
Flag any unverifiable sources.

Expected output:

  • Citation format compliance report
  • Unverifiable citations flagged
  • Suggested corrections

Literature Review Mode

Use $academic-research-suite: ars-lit-review

Topic: AI ethics in higher education
Scope: 2020-2026, peer-reviewed only
Output: thematic synthesis with citation matrix

Paper Formatting and Conversion

Format Conversion

Use $academic-research-suite: ars-format-convert

Input: [attach paper.md]
From: Markdown
To: LaTeX (IEEE conference template)
Preserve: citations, figures, tables

AI Disclosure Statement

Use $academic-research-suite: ars-disclosure

Paper stage: final draft before submission
AI tools used: Codex for literature search, outline generation, citation formatting
Human contribution: research design, analysis, interpretation, revision
Output: journal-ready disclosure statement

Revision Workflow

Revision Coach

For guidance without executing changes:

Use $academic-research-suite: ars-revision-coach

[attach draft.md and reviewer_comments.txt]

Analyze reviewer feedback and suggest revision strategy.
Prioritize major vs minor revisions.
Do not rewrite yet - provide plan only.

Execute Revisions

Use $academic-research-suite: ars-revision

[attach draft.md and reviewer_comments.txt]

Execute revisions based on Reviewer 2 methodology concerns.
Track changes: show original -> revised for each section.
Update Material Passport with revision metadata.

Troubleshooting

Skill Not Found

If /skills doesn't show academic-research-suite:

  1. Verify installation path: ls -la "$HOME/.codex/skills/academic-research-suite"
  2. Reinstall using update command above
  3. Open a new Codex conversation (old sessions cache skills)

Multiple ARS Skills Showing

If you see academic-paper, deep-research, etc. as separate skills:

# Remove all ARS skills
rm -rf "$HOME/.codex/skills/academic-"*
rm -rf "$HOME/.codex/skills/deep-research"
rm -rf "$HOME/.codex/skills/experiment-agent"

# Reinstall single suite
python "$HOME/.codex/skills/.system/skill-installer/scripts/install-skill-from-github.py" \
  --repo Imbad0202/academic-research-skills-codex \
  --ref main \
  --path skills/academic-research-suite \
  --method git

Wrong Workflow Selected

If ARS routes to the wrong workflow, be more explicit:

Use $academic-research-suite deep-research workflow in socratic mode.

[your request]

Or use specific alias:

Use $academic-research-suite: ars-outline

[your request]

Citation Verification Fails

When citations cannot be verified:

  • ARS should mark them as unverified rather than inventing support
  • If verification is critical, provide DOIs or URLs for sources
  • Use ars-citation-check mode with source files attached

Codex Warnings

These warnings are non-blocking:

  • [features].codex_hooks is deprecated - update Codex config when convenient
  • hooks need review before they can run - ARS Codex doesn't require hooks for normal use

Advanced Patterns

Staged Pipeline with Checkpoints

Use $academic-research-suite academic-pipeline workflow.

Stage 0: Intake and dashboard
STOP and show dashboard for approval

[After approval]

Stage 1: Research question refinement via Socratic dialogue
STOP and show refined question

[Continue stage by stage]

Parallel Literature Search

Use $academic-research-suite deep-research workflow.

Search strategy: parallel tracks
Track 1: Scopus, Web of Science - quantitative studies
Track 2: Google Scholar - grey literature, policy docs
Track 3: Specialized databases - ERIC, ProQuest Education

Synthesize results into single literature matrix.

Multi-Pass Review

Use $academic-research-suite academic-paper-reviewer workflow.

Pass 1: Structure and contribution (desk-reject check)
[Wait for my review]

Pass 2: Methodology rigor
[Wait for my review]

Pass 3: Citation integrity and ethics
[Final decision]

Integration with Research Tools

With Reference Managers

Use $academic-research-suite: ars-citation-check

Citation library: Zotero export attached (bibtex format)
Paper: [attach draft.md]

Cross-check all in-text citations against Zotero library.
Flag missing entries and formatting errors.

With Data Analysis Code

Use $academic-research-suite experiment-agent workflow.

Existing analysis: [attach analysis.py, results.csv]
Task: Create reproducible experiment protocol around this analysis
Output: protocol.md, requirements.txt, reproduction checklist

With Institutional Templates

Use $academic-research-suite: ars-format-convert

Input: [attach paper.md]
Template: University PhD thesis LaTeX template [attach template.tex]
Preserve: custom environments for theorems, university citation style

CLI Smoke Test

Test routing without full execution:

codex exec --ephemeral --sandbox read-only \
  -C /path/to/academic-research-skills-codex \
  'Use $academic-research-suite. Router smoke test only. Classify workflow and mode for: I want to write a paper on AI adoption but do not have a research question yet.'

Expected: Route to deep-research socratic mode.

Key Differences from Claude Code Version

This Codex package differs from the original Claude Code ARS:

Feature Claude Code Codex Package
Installation Plugin marketplace Git-based skill installer
Commands Native /ars-* slash commands Emulated aliases via router
Agent teams Automatic background agents Inline unless explicitly delegated
Hooks SessionStart, SubagentStop Vendored for traceability only
Cross-model review GPT/Gemini secondary dispatch Anthropic Claude API (explicit config)
Updates Marketplace auto-update Manual reinstall

Version Information

Additional Resources

For issues specific to the Codex packaging, open an issue at the Codex package repository. For ARS workflow issues, consult the original Claude Code repository.

Related skills

researchmattpocock575KInvestigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.paper-context-resolverlllllllama451KRigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing Renv-and-assets-bootstraplllllllama450KRigor Setup skill for README-first deep learning repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.ai-research-explorelllllllama311KRigor Explore compatible skill slug for meaningful and potentially novel deep learning research candidates. Use when the researcher has chosen the task family, dataset, benchmark, evaluation method, provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, fair comparison, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow c

Search skills and MCP servers

Fuzzy search across 23,137 skills and servers