Scholar Evaluation

What it does

Delivers developmental feedback on a scholarly work — paper, draft, protocol, literature synthesis or research idea — leading with qualitative judgment; optional scores only map submitted evidence onto a predeclared bounded rubric.

• Forces you to define the construct before writing criteria.
• Ships offline Python 3.11 CLIs (no network, no credentials, no subprocesses) for rubric validation, bounded scoring, evidence traceability, inter-rater agreement, weight sensitivity and a fail-closed process checklist.
• Bans quality inference from impact factor, h-index, citation counts, venue or institutional prestige.
• Keeps missing and not_applicable distinct from zero, and observation distinct from interpretation.

⚠️ Explicitly prohibited for hiring, promotion, admissions, funding, awards or discipline. It never ranks people.

Who it's for

• Journal/conference reviewers, thesis advisors and lab-internal review leads.
• Research groups that want a repeatable, documented critique workflow.
• Staff auditing whether a low-stakes assessment process documents provenance, uncertainty, fairness and appeals.

Examples

1. Pre-submission self-check: break your draft down criterion by criterion, log strengths, limitations and non-prescriptive improvement options with evidence locators.
2. Group review: have three reviewers rate independently, then run summarize_agreement.py on the CSV to surface disagreements before a calibration meeting.
3. Process audit: run check_process.py against a departmental seminar review to confirm appeal routes, accessibility accommodations and retention policy are documented.

Install the skill I found on Claude Skill Mart.
Copy the skills/scholar-evaluation folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/scholar-evaluation/.
When it's done, tell me in one line what this skill can do.