Prompt Library Gardener
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md and install @user_15292d5a/yjkj-prompt-library-gardener.
About this skill
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.
Use Cases
- A content editor pastes 30 writing prompts and asks for job, output, and audience tags plus duplicate merging.
- An engineering lead organizes code-review and debugging prompts into names, tags, defaults, and reuse notes.
- An ops manager consolidates meeting-summary, email, and training prompts into a searchable task-based index.
- An AI coach catalogs student-submitted prompts, flags testing candidates, and marks archive options.
Best For
- Content editors maintaining a prompt library who need scattered writing templates turned into a searchable index.
- Engineers managing agent workflows who need task and output deduplication with documented reuse notes.
- Operations owners of team AI tooling who need prompts tagged by audience and a default version per group.
- Product managers organizing personal knowledge bases who need similar variants merged into a naming convention.
Related Skills
Model routing, persistent parameter management, self-check repair, and global default model control for XiaoYi Claw.
Install, update, and manage OpenClaw Skills through SkillHub, with automatic detection after installation.
Provides API endpoints for AI agents to post bottles and graffiti, browse the feed, and interact with likes and comments.
ReqPlan constrains agent development, debugging, and analysis workflows with a seven-stage state machine, checkpoints, quality audits, and local Harness artifacts.