Prompt Library Gardener

Problem

When prompts are scattered across notes, spreadsheets, or chat logs, the practical issue is retrieval: you know related prompts exist but cannot pick the best version for the current task. Similar prompts accumulate without clear names, tags, or ownership, so users rewrite from memory or keep duplicate variants. prompt-library-gardener addresses organizing an existing prompt library, not generating prompts from scratch.

How It Works

It works only from prompt text and context provided in the conversation. It does not scan local folders, note apps, browser history, cloud drives, email, or private workspaces. The workflow is engineering-like: create an inventory with ID, purpose, output type, and status; group prompts by the job the user wants to complete, such as research, writing, code, decisions, or communication; classify overlap into exact duplicate, near duplicate, useful variant, legacy version, and items needing user decisions; then produce names, tags, and reuse notes. Names usually follow a job - output - constraint pattern, and tags are kept to 3 to 7 useful items, avoiding broad terms like prompt or AI. For sensitive prompts, it recommends replacing confidential details with placeholders and preserving original intent instead of silently discarding variants.

Boundaries

Use it for cleaning an existing library, building an index, merging near-duplicates, and documenting usage. Do not use it for deep optimization of a single prompt, designing a workflow from zero, or asking it to pull prompts from local systems. For large libraries, provide prompts in labeled batches. If a complex taxonomy feels necessary, start with job and output dimensions first, because over-categorization can make retrieval slower rather than faster.