Skip to content

Commit 49aacc2

Browse files
author
Vambrocop
committed
Add academic research suite integration note
1 parent 796eb48 commit 49aacc2

1 file changed

Lines changed: 120 additions & 0 deletions

File tree

Lines changed: 120 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,120 @@
1+
# Academic Research Suite Integration Note
2+
3+
External repositories:
4+
5+
- Claude Code version: https://github.com/Imbad0202/academic-research-skills
6+
- Codex-native version: https://github.com/Imbad0202/academic-research-skills-codex
7+
8+
Local status:
9+
10+
```text
11+
reviewed and installed into local Codex skills on 2026-05-20
12+
installed skill: academic-research-suite
13+
license: CC BY-NC 4.0
14+
```
15+
16+
Codex usually needs a restart before newly installed skills appear in the active skill list.
17+
18+
## What It Is
19+
20+
Academic Research Suite (`ARS`) is a broad research-to-publication workflow suite. The Claude Code repository contains four main skills:
21+
22+
```text
23+
deep-research
24+
academic-paper
25+
academic-paper-reviewer
26+
academic-pipeline
27+
```
28+
29+
The Codex-native repository packages these into one skill:
30+
31+
```text
32+
academic-research-suite
33+
```
34+
35+
It covers:
36+
37+
- Socratic research-question refinement;
38+
- literature review and systematic review planning;
39+
- paper outlining and drafting;
40+
- citation and integrity checks;
41+
- simulated peer review;
42+
- end-to-end research-to-paper pipeline;
43+
- experiment planning and validation.
44+
45+
## Why It Matters For EvidenceForge
46+
47+
EvidenceForge is narrower and more method-specific:
48+
49+
```text
50+
systematic review
51+
meta-analysis
52+
umbrella review
53+
evidence synthesis reproducibility
54+
coding sheets
55+
effect-size logic
56+
high-value article paradigms
57+
```
58+
59+
ARS is useful as a broader orchestration and mentoring layer, especially when the user starts with a vague idea rather than a defined review protocol.
60+
61+
## Recommended Division Of Labor
62+
63+
| Task | Primary workflow |
64+
| --- | --- |
65+
| vague topic, research-question narrowing | `academic-research-suite` Socratic / deep-research |
66+
| systematic review protocol, PECO/PICO, coding schema | EvidenceForge |
67+
| meta-analysis execution logic and effect-size validation | EvidenceForge / `meta-analysis-forge` |
68+
| full paper outline and drafting | `academic-research-suite` or `nature-writing`, with EvidenceForge evidence ledger |
69+
| manuscript peer-review simulation | `academic-research-suite` reviewer |
70+
| high-impact article reproducibility template | EvidenceForge |
71+
| final polishing, paper-to-PPT, Nature-style figures | `nature-skills` |
72+
73+
## Installation Commands
74+
75+
For Codex:
76+
77+
```powershell
78+
python C:\Users\Vambr\.codex\skills\.system\skill-installer\scripts\install-skill-from-github.py `
79+
--repo Imbad0202/academic-research-skills-codex `
80+
--ref main `
81+
--path skills/academic-research-suite `
82+
--method git
83+
```
84+
85+
For Claude Code:
86+
87+
```text
88+
/plugin marketplace add Imbad0202/academic-research-skills
89+
/plugin install academic-research-skills
90+
```
91+
92+
## Governance Rule
93+
94+
Do not vendor the full ARS repositories into EvidenceForge by default.
95+
96+
Reasons:
97+
98+
- the license is CC BY-NC 4.0, so reuse is best kept as linked external integration unless terms are checked for the intended use;
99+
- ARS is broad and can blur EvidenceForge's evidence-synthesis boundary;
100+
- wholesale copying creates maintenance burden;
101+
- EvidenceForge should preserve its own stricter meta-analysis and reproducibility rules.
102+
103+
Preferred approach:
104+
105+
```text
106+
install externally
107+
link source
108+
use ARS for broad orchestration
109+
use EvidenceForge for evidence-synthesis decisions
110+
adapt only small, audited process rules when needed
111+
```
112+
113+
## Use Prompt
114+
115+
```text
116+
Use academic-research-suite to help narrow the research question and produce a staged research plan.
117+
Then switch to EvidenceForge for the systematic-review protocol, coding sheet, effect-size logic,
118+
and reproducibility checklist.
119+
Do not let a broad writing pipeline override EvidenceForge's evidence-synthesis guardrails.
120+
```

0 commit comments

Comments
 (0)