/rubric

name: rubric
description: Turn an assignment prompt into a ready-to-use grading rubric (a.k.a. scoring guide / grading guide) with zero blank cells. USE THIS SKILL when a teacher, professor, TA, or parent says "I need a rubric for…", "make me a scoring guide", "how should I grade this?", "build an analytic rubric", "holistic rubric for a narrative", "4-point mastery rubric", "rubric aligned to Bloom's / DOK / SOLO / CCSS / NGSS", "turn this assignment into a rubric", "student-facing rubric I can hand out", "grading criteria for [project]", "help me grade this essay", "what should I look for when grading", or "rubric for my kid's homeschool project". Handles K-2 through graduate, analytic or holistic, points or qualitative. Outputs markdown table + CSV with fully populated descriptors in every cell.
license: MIT

rubric

Build defensible, fully-populated grading rubrics from an assignment prompt. Holistic or analytic, 3-6 criteria x 3-5 levels, aligned to Bloom's (default), DOK, SOLO, or standards. Every cell is written — no "TBD", no "same as above", no cells that are just "less of the previous level."

When to trigger

Fire on any phrase combining "rubric / scoring guide / grading guide / criteria" with an assignment, or on any of these cues. Teachers rarely say "rubric" — cover the synonyms.

• "I need a rubric for [assignment]."
• "Make me a rubric for this essay."
• "Grade this essay assignment" / "help me grade…"
• "Build a scoring guide for [project]."
• "Write an analytic rubric for a lab report."
• "Holistic rubric for a 6th-grade narrative."
• "4-point mastery rubric for…"
• "Rubric aligned to DOK" / "aligned to Bloom's" / "aligned to CCSS."
• "Scoring criteria for a group project."
• "How should I grade this?"
• "Turn this assignment prompt into a rubric."
• "Make a student-facing rubric I can hand out."
• "What should I look for when grading [X]?"

Inputs

**Required:**

• **Assignment prompt** — pasted text, uploaded PDF/DOCX, photo of a handout, or URL to a class page.

**Optional** (ask only if ambiguous after inference; otherwise use defaults from § Grade-band defaults):

• `grade` — K-2 | 3-5 | 6-8 | 9-12 | college | grad
• `subject` — ELA | math | science | social studies | CS | art | PE | world language | other
• `rubric_type` — `holistic` | `analytic`
• `criteria_count` — 3-6 (default 4)
• `levels_count` — 3-5 (default 4)
• `alignment` — `bloom` (default) | `dok` | `solo` | `ccss` | `ngss` | `state:<code>` | `ib` | `none`
• `scoring_style` — `points` | `qualitative` | `hybrid`
• `total_points` — integer (default 100 when `scoring_style=points`)
• `audience` — `teacher` (default) | `student` | `both`
• `outputs` — any subset of {`markdown`, `csv`, `classroom_csv`, `student_handout`}

Inferring grade / assignment-type from phrasing (before asking)

Infer first; ask ONLY if you cannot confidently pin down the grade band AND the assignment type.

| Cue in user's phrasing | Likely grade band | Likely assignment type | |---|---|---| | "my kindergartener," "my 5yo," "picture rubric," "stars or smileys" | K-2 | performance / drawing / letter formation | | "my 3rd grader," "book report," "show your work" | 3-5 | short essay / math problem set | | "middle school," "5-paragraph essay," "historical figure project" | 6-8 | essay / group project / lab | | "AP ___," "thesis statement," "persuasive essay," "MLA cited" | 9-12 | essay / lab report / research paper | | "syllabus," "semester," "final project," "TA," "office hours" | college | capstone / coding project / paper | | "dissertation," "proposal," "IRB," "cohort" | grad | thesis / proposal (narrative holistic) | | "state test," "STAAR," "NWEA," "performance task" | — | prefer DOK alignment | | "CCSS," "Common Core," "W.9-10.1" | — | prefer CCSS tags | | "NGSS," "SEP," "phenomenon," "3-dimensional" | — | prefer NGSS SEP tags | | "group project," "team," "collaborative" | — | add Individual Contribution + Collaboration criteria | | "my ELL / newcomer / bilingual student" | — | mirror language; add plain-language student version | | "my kid's homeschool" / "for my own kid" | use stated grade | grade-appropriate; home-friendly |

If grade is still unclear, ask ONE clarifying question in this exact shape:

> Quick check so I can calibrate this — which grade is this for? (e.g. **K · 3rd · 7th · 10th · college**)

Do not stack questions. Do not ask about alignment, levels, or points unless the user explicitly asks for tight customization.

Workflow

1. **Ingest assignment prompt.** Accept text, URL (fetch), PDF/image (extract). If ingestion was non-trivial, confirm the extracted prompt back in 2-3 lines. 2. **Identify the task type.** Essay, lab report, project, presentation, coding assignment, portfolio, performance task, etc. Note the verbs used in the prompt ("analyze," "design," "compare," "evaluate"). 3. **Pick taxonomy.** Default `bloom`. If the prompt mentions assessment / state testing / performance task, or the subject is math/science, prefer `dok`. If user supplied `alignment`, honor it. See `references/taxonomies.md` for verb banks + decision tree. 4. **Propose 3-6 criteria.** Observable, non-overlapping, one-trait-per-criterion. See `references/criterion-writing.md` for rules and criterion-set templates by assignment type. 5. **Generate level descriptors.** For each criterion x level cell, write a concrete, measurable descriptor. Use parallel phrasing + one quantifier ladder per row. No empty cells; no "same as above"; no "less of the previous." See `references/descriptor-phrasing.md`. 6. **Align & label.** Attach a taxonomy/standard tag per criterion, e.g. `*Analyze — Bloom's 4*` or `*CCSS.ELA-Literacy.W.9-10.1*`. For standards, see `references/standards-crosswalk.md`. 7. **Weight and score.** Default equal weights. If `scoring_style=points`, distribute `total_points` by weight, round to integers, and verify: (a) level points within each criterion are monotonically increasing, (b) criterion totals sum