Finer Criteria Check

SOP: check research-question quality against each of the 5 FINER criteria

yogsoth-ai Updated 512 repo stars
No DARE runtime: 271 SKILL.md files executed by any capable agent, instead of a seven-layer framework stack

File contents

FINER Criteria Check

FINER 5-criteria per-item check — the standardized test of research-question quality.

HARD-GATE

Pipeline

  1. Pre-check: is the RQ a complete sentence
  2. F (Feasible): is it feasible under the available resources/time/data
  3. I (Interesting): is it interesting to scholars and practitioners in the field
  4. N (Novel): does it provide new knowledge/new perspective/new method
  5. E (Ethical): does it comply with research ethics
  6. R (Relevant): is it relevant to the field/society/practice
  7. Overall verdict: all pass / partial pass
  8. Revision suggestions: give specific revision directions for failed items
  9. Output: the 5 verdicts + revision suggestions (if any)

Output Format

F: PASS/FAIL — [rationale]
I: PASS/FAIL — [rationale]
N: PASS/FAIL — [rationale]
E: PASS/FAIL — [rationale]
R: PASS/FAIL — [rationale]

Overall: PASS (5/5) / PARTIAL (X/5)
Suggestions: [specific revision suggestions for FAIL items]

yogsoth-ai/de-anthropocentric-research-engine commit 26905c3732

Frequently asked questions

npx get-skill add yogsoth-ai/finer-criteria-check