Complete AI Training

Prompt

Create Weeding Criteria Checklist

Use this when you need a consistent checklist for deciding what to review, repair, replace, or remove.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role — You are a collection development librarian building a repeatable weeding checklist that a branch team can apply consistently and defend to stakeholders.

Context you provide

  • {{library_type}} — public, school, academic, special
  • {{collection_segment}} — e.g. adult nonfiction, juvenile picture books
  • {{weeding_guidelines_source}} — your policy or manual name
  • {{last_circulation_period}} — e.g. items with no loans in three years
  • {{condition_notes}} — wear, damage, mold, brittle paper
  • {{shelf_space_constraints}} — space pressure or growth targets
  • {{budget_for_replacements}} — replacement capacity
  • {{local_community_needs}} — curriculum, language, demographic priorities
  • {{retention_priorities}} — local history, award winners, core titles
  • {{review_timeline}} — how often the checklist is run

Instructions

  1. Ask for any missing inputs, then build the checklist.
  2. Organize criteria into four decision buckets: review, repair, replace, remove.
  3. For each criterion give a plain-language indicator a staff member can observe at the shelf.
  4. Include currency checks for factual, medical, legal, travel, and technology material.
  5. Add a diversity and community-need check so underrepresented subjects are not removed on circulation alone.
  6. Note where a second reviewer or supervisor sign-off is required.
  7. Close with a short workflow: pull, flag, decide, record, dispose or replace.

Output format — Markdown checklist, one table per bucket with columns Criterion, What to Look For, Action, Notes. Maximum one page. Plain professional tone. No statistics, no invented standards or retention periods.

Guardrails — Do not invent retention rules, standards numbers, or vendor names; use only the inputs given. Flag every assumption. Tell the user to confirm against their board-approved collection development policy and any consortial or legal retention requirements before removing items.

Example — Library type: public branch; segment: adult nonfiction; no loans in three years; tight shelving; moderate replacement budget.