prompt-librarylisted

# prompt-library Overhaul task 9. Reusable prompts (e.g. a very detailed feature-update brief; a "design the whole X" prompt that transfers across sibling projects; a manual/onboarding prompt) get curated once, de-privacy'd, tagged by scenario, and made easy for a human to find and for an agent to reuse. ## Routing: check the library before writing a prompt from scratch When a request matches a scenario the library already covers, READ THE ENTRY FIRST and adapt its Optimized block instead of composing a new prompt. Run `plib.py find --query "<the user's words>"`, or scan the `By scenario` section of `recommendations/prompt-library/INDEX.md`. Scenarios currently covered include feature specs and multi-part update briefs, skill authoring, documentation (subsystem docs, completeness audits, in-app manuals and onboarding), deployment and infrastructure, data-platform and schema work, research and experiment design, bug triage, reporting, and proposal writing. Two rules when you reuse one. Read `When NOT to use` before adopting an entry — several of them are actively wrong for the neighbouring case, and that section is where the judgement lives. And treat the `Optimized` block as a template with `<placeholders>` to fill, not as text to paste unchanged. ## Use - **Save** a prompt — the ORIGINAL goes on stdin, the rest as flags. The privacy gate runs over the whole rendered document first and REFUSES if anything still looks private: ``` python3 scripts/plib.py add --titl