Powerofjinbo
- 32 repo stars
- Updated July 4, 2026
- powerofjinbo/phdtaketaketake
phdtaketaketake is a Claude Code skill packaged as a marketplace plugin that scores applicant profiles and ranks candidate advisors using connection-first, evidence-traced signals so users can prioritize PhD applications
About
phdtaketaketake is a Claude Code skill and marketplace plugin that helps STEM applicants find and rank PhD advisors using web-verified connection graphs and deterministic scoring scripts. Use it when you need to triage US graduate programs, compare PIs against your profile, or compile a CV—not when you are building production application code.
- Connection-first ranking emphasizing advisor network edges over h-index alone
- Python CLIs: discovery-plan, collect-evidence, audit, match, CV compile
- Strict evidence rules—missing data widens confidence bands instead of guessing
- npx installer for Claude Code, Codex, and Cursor project rules
Powerofjinbo by the numbers
- Data as of Jul 26, 2026 (Skillselion catalog sync)
/plugin marketplace add powerofjinbo/phdtaketaketakeAdd your badge
Show developers this marketplace is listed on Skillselion. Paste this into your README.
| repo stars | ★ 32 |
|---|---|
| Last updated | July 4, 2026 |
| Repository | powerofjinbo/phdtaketaketake ↗ |
How do you rank PhD advisors and application strength from real publication and collaboration evidence instead of guesswork or raw h-index?
Rank PhD advisor targets with evidence-first connection scoring and bundled CLIs fed by OpenAlex, PubMed, and Semantic Scholar.
Who is it for?
Software engineers and STEM applicants targeting US PhD programs—especially physics/HEP and materials science—who want advisor shortlists grounded in fetched sources.
Skip if: Hiring managers, industry job search, or anyone not planning graduate-school applications.
What you get
The agent builds a profile, collects cited evidence, runs match.py for ranked advisors with confidence bands, and can export schemas or compile LaTeX CV PDFs.
Plugins in this marketplace
1 plugin - install individually after you add the marketplace.
Recommended Marketplaces
FAQ
Is the score an admission probability?
No—it is a 4.0-scale relative application-strength index with explicit design boundaries; missing sources widen confidence rather than being imputed.
Can the agent invent collaboration edges?
No—every edge and fact must trace to a fetched source; guessing or training-memory inference is forbidden and strict-evidence mode can hard-fail unsourced runs.
How do I install for Claude Code?
Run npx @Powerofjinbo/phdtaketaketake install --claude or use marketplace add/install, then pip install -e the printed package path and run npx @Powerofjinbo/phdtaketaketake doctor.