home/writing/cot-v4

Cot v4

GPTClaudeGemini··1,191 copies·updated 2026-07-14
cot-v4.prompt
Classify this essay's CEFR level through structured analysis. IMPORTANT: Essay length is NOT a proficiency indicator.

REASONING PROTOCOL:

Step 1 - MORPHOSYNTAX (Don't assume length = complexity):
- What verb forms appear? (simple vs. complex tenses, modals, conditionals)
- What sentence structures? (simple, compound, complex subordination)
- Is complexity present regardless of essay length?

Step 2 - LEXICAL SOPHISTICATION (Not vocabulary size):
- Academic/abstract terms used?
- Precise lexical choice vs. generic terms?
- Idiomatic or colloquial expressions?

Step 3 - DISCOURSE FEATURES (Not just organization):
- Cohesion: explicit connectors (B1) vs. implicit flow (C1)?
- Argumentation: listing (B1) vs. developed reasoning (B2+)?
- Register: consistent formal/informal control?

Step 4 - DIAGNOSTIC MARKERS (Anti-bias check):
- B2 MUST have: complex subordination + hypotheticals + abstraction
- C1 MUST have: register control + sophisticated argument + discourse cohesion
- C2 MUST have: idiomatic fluency + pragmatic nuance
- Avoid B1 default if these features present!

Step 5 - LEVEL DECISION (Feature-based):
If uncertain between adjacent levels, choose HIGHER if features present.
Short essays with dense complex features = advanced proficiency.
Long essays with simple features = intermediate proficiency.

Essay:
{essay_text}

After reasoning, output ONLY the final level: A2, B1, B2, C1, or C2

fill the variables

This prompt has 1 variable. Pro fills them into a ready-to-paste prompt for you — no manual find-and-replace.

{essay_text}
Unlock with Pro →

when to use it

Community prompt sourced from the open-source GitHub repo SiemonCha/ECM3401-LLM-Essay-Scoring (MIT). A "Cot v4" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

writingcommunitygeneral

source

SiemonCha/ECM3401-LLM-Essay-Scoring · MIT