Library

"Builds a prompt-first home page from HubSpot for a rep who works through Claude instead of the HubSpot screens: today's meetings with briefs, waiting email and deals, where every card and every \"log it\" button is a prompt to copy into Claude. Use when a rep won't learn new software but uses Claude, or to recreate the Claude-first rep page from the HubSpot demo."

"Checks every open deal's stage against its actual emails, meetings, calls, notes and tasks in HubSpot, and lists the deals whose stage doesn't match, each with a suggested stage, a confidence and the quote that shows it. Writes nothing. Use for a weekly pipeline review, before a forecast call, or when stages drift because reps update them last."

"Checks HubSpot's official developer docs and changelog for what changed since the last check in the tools you use (the remote MCP server and Claude connector, the Agent CLI, CRM, marketing email and import APIs), and writes it into a notes file your agent reads before any HubSpot work. Use weekly, or before building anything on HubSpot."

"Builds a newsletter performance dashboard from HubSpot marketing email data: opens, clicks and delivered per send, by publication, week over week, against your own averages, with every chart also shown as a table. Use when a marketer wants a weekly newsletter report without building HubSpot reports, or to recreate the marketing dashboard from the HubSpot demo."

"Builds this week's newsletter as a HubSpot marketing email draft by cloning last week's send and swapping in the stories and ads you give it, then checks links and settings and stops before sending. Use when a newsletter is assembled from the same template every week and a person still reviews and sends it."

"Audits a HubSpot portal read-only and builds a scorecard page: where contacts came from, lifecycle stages, companies missing domain or industry, deals with no amount, past close dates or no contact, pipeline stages in use, workflows switched on, property sprawl and lists, each with the fix and a link to HubSpot's docs. Use when taking over a portal, after a migration, or before planning automation."

"Writes a pre-meeting brief for a client meeting from HubSpot alone: top line, what they bought by year, what's new since the last contact, the deal and last email, who's in the room, what to pitch, agenda, questions, likely objections and gaps. Use before a client meeting, or each morning for every meeting today and tomorrow, or to recreate the pre-meeting brief from the HubSpot demo."

"Writes eight ready-to-paste Claude prompts for one person's day, filled with real names, companies, deals and dates from their HubSpot data and grouped by job (follow up, prep, renew, clean up, report). Use each morning to give someone prompts they can run without writing one, or to add the prompt recommendations to a morning page."

"Lists the companies that bought this year (or last year) and have nothing booked for next year, from HubSpot deals, warmest first, with amounts, owner, last client email and a prompt to start each conversation. Use in renewal season or when a rep asks who to renew next."

"Builds a weekly rep activity page from HubSpot: emails, client replies, meetings, calls, notes and tasks each rep logged this week against the last eight, open pipeline, quiet deals and meetings with no deal. Use when a manager wants a weekly team summary that reps can read without logging in to HubSpot."

"Builds a sales rep's morning page from HubSpot alone: one HTML file with today's next moves, meetings, client email waiting on a reply, renewals, open deals and ready-to-copy Claude prompts. Use when a rep wants a daily \"what do I do first today\" page from their own HubSpot data, or to recreate the simple rep page from the HubSpot demo."

"Builds a sales manager's team page from HubSpot: booked revenue against the goal by rep, what the team logged this week, meetings ahead, client email waiting on the team, the deals table with owners, deals whose stage doesn't match their activity, and Ask Claude prompts, with a picker to view any rep. Use when a sales manager wants one daily team view, or to recreate the manager page from the HubSpot demo."

Scores every human prompt saved in a repo's git history against a five-part rubric and writes a report of the three habits to fix first, what to keep doing, and rewrites of the weakest prompts using real examples. Use when someone asks how to prompt better, wants tips on their prompting, asks for a prompt report, review, or audit, or wants to see whether their prompts are improving over time.

Version, cite, evaluate, and distribute prompts like code.

Use whenever a skill or agent rule (any SKILL.md) is edited, improved, fixed, created, removed, or receives feedback.

Keeps third-party agent skills safe and reproducible in a repo — read before install, pin to commit SHAs, commit the lockfile, and re-review on update. Use when installing, updating, or auditing agent skills from any external source in any project.

Evaluates a skill distributed on uplift.page using live platform data — conversion cleanliness, size, citation, freshness, workspace usage, and outcome feedback. Use when asked whether a skill is healthy, worth keeping, or needs work, when auditing a workspace library, or before promoting a skill to the public library.

Manages a workspace's skill library on uplift.page — add, update, pull, install, and archive skills through the uplift MCP workspace tools. Use when asked to save a skill to uplift.page, install or pull a skill from uplift.page, update or remove a distributed skill, or sync a project's skills with a workspace library.

Curates the prompts served by the uplift.page prompt API — audits against best practices, applies minor fixes through the API, and files change requests for major rewrites. Use when asked to review, improve, fix, or update prompts in the public library or a workspace library.

Offers to save newly created or edited agent skills to the uplift.page workspace library so they are versioned and distributed. Use whenever a skill or agent rule file (any SKILL.md or equivalent) is created, edited, fixed, or improved in any project.

Builds a non-technical how-to page for any app or repo, from install to full understanding, with real screenshots, step-storyboard Remotion demos, page-wide search, redaction of real people, a currency check, a password-gated static deploy, and the CLAUDE.md rules that keep it current. Use when asked for a user guide, how-to, onboarding walkthrough, help page or instructions page for an app, product or internal tool.

Matches every user-facing string — email subjects, notifications, buttons, errors, empty states — to the product's vibe and the reader it is written for. Use when writing or reviewing any copy a user will see.

Pulls the current documentation for any library, API, app, or vendor into the repo as short dated notes agents read before working, and keeps them fresh with a weekly check that flags which sources changed, starting from the changelog. Use when setting up AI context for a tool or vendor, when the model's knowledge of a fast-moving product may be stale, when asked to refresh, sync, or check docs, or before relying on a vendor's API, CLI, or limits.

Writes short, human-legible READMEs that lead with a quick start and surface the manual steps the human still needs to do. Use when creating or updating a README.

Docs site with quick starts, task guides, and generated reference.

Builds a customized training document for one reader — a how-to guide, onboarding walkthrough, study course, or app manual — from what that reader already uses and what they want to be able to do, about any topic, tool, or repository. Asks only the intake questions the context leaves open, then delivers one of three formats — a single-file HTML guide, a keyboard-navigable slide deck, or a standalone searchable Next.js docs site. Use whenever someone asks to teach, train, onboard, explain, document how to use, write a tutorial or course, or make learning material for a person or team, even if they never say "training".

Diagnoses bugs by reproducing first and naming the root cause before any code changes — reproduce, instrument, bisect, prove the fix against the reproduction, leave a regression test. Use whenever something is reported broken, failing, or behaving wrong, before attempting a fix.

Interrogates a finished diff adversarially before it becomes a commit or PR — correctness, edge cases, security, and performance challenges the author must answer or fix. Use when a feature or fix is declared done, before committing, opening a PR, or handing work to review.

Enforces Uplift Duo house code style — naming, comments, error handling, and structure. Use when writing, editing, or reviewing code in any language.

Produces dual effort estimates — traditional human-developer weeks alongside an agent-driven build-days + tokens + cost figure assuming a tight, increasingly-unsupervised feedback loop. Use when asked to estimate effort, time, cost, timeline, "how long", "how much", or scope for a build, feature, migration, or refactor.

Applies the Uplift Duo git workflow — branch naming, commit triggers, the emoji commit message format, GitHub defaults, and merge rules. Use when branching, committing, merging, or creating repos.

Audits a directory of git repositories and produces an interactive Project Audit — per-repo descriptions, % complete, ratings (market size, value, code quality, security, momentum), bugs by severity, feature suggestions and improvements with type tags, major-change timelines, tech-stack analysis with recommended alternatives, AI model attribution by percent, token usage (commit-block estimates vs measured transcripts with a completeness comparison), total AI thinking time, and a copy-paste improvement prompt per repo. Use when the user asks to audit, review, rate, or take stock of their repos, projects, or portfolio, asks "what's the state of my projects", or wants a project audit for any time window.

Verifies UI changes in a real browser before declaring them done — drive the changed flow, check the console, test the mobile viewport, and report what was actually seen. Use whenever a change touches anything a user sees or clicks, and before any "fixed" or "done" claim about frontend work.

Hands work to the human only after verifying everything the agent can itself, then gives a short, risk-ordered list of the checks only a person can do (real devices, their accounts and inboxes, live payments, production, taste and business judgment), each with exact steps, the expected result, and why it needs a human. Use after any major code push, finished feature, opened PR, merge, or deploy, and whenever about to ask the user to test, try, check, confirm, or "let me know if it works".

Vitest + Playwright + CI gates: fast unit core, few real E2E flows.

Gets verification done in apps that sit behind sign-in — via the app's own dev bypass, a temporary identity, or the user's already-signed-in browser. Apply when a login wall blocks opening, testing, or screenshotting an app, or when about to skip browser verification because the app requires authentication.

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

A tool-using agent with typed tools, MCP, and an eval harness.

Appends two or three sharp self-check questions after substantive advice, plans, estimates, or drafts so the reader pressure-tests the work instead of rubber-stamping it. Use once per conversation, after delivering a recommendation, estimate, plan, or draft someone will act on.

House conventions for designing MCP servers and tools — tenant-from-credential auth, instructive errors, tiered access, and tool descriptions written for the calling model. Use when building, extending, or reviewing any MCP server or its tool definitions.

Connects any agent or client workspace to the uplift.page prompt platform — MCP server setup, workspace keys, and the plain-HTTP fallback. Use when setting up uplift.page access, onboarding a client workspace, wiring the uplift MCP server into an agent, or when a workspace tool reports a missing or invalid key.

Tokens, primitives, and motion rules - shadcn/ui + Tailwind + Radix.

Anti-slop rules for landing pages, portfolios and marketing-site redesigns. Read the brief, set three dials (variance, motion, density), and avoid the templated AI look. Use when designing or redesigning a landing page, portfolio, or editorial site; not for dashboards or dense product UI.

Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.

Clean, minimalist, developer-forward UI: the design prompt behind uplift.page.

Three.js + React Three Fiber + Rapier physics for browser-native 3D.

Ship a polished 2D web game with Phaser, TypeScript, and Vite.

48-hour jam plan: KAPLAY, one mechanic, ruthless scope.

Realtime multiplayer in the browser with PartyKit rooms and client prediction.

Next.js App Router + Supabase + Tailwind: the boring stack that ships.

Migrates a Next.js App Router app to explicit caching — enabling cacheComponents and partialPrefetching, auditing implicit-caching assumptions, adding 'use cache' and Suspense shells, and locking behavior with navigation tests. Use when upgrading an app to Next.js 16.x or enabling Instant Navigations on an existing project.

Applies Uplift Duo Next.js conventions — App Router structure, Tailwind, Motion for React, explicit caching, server actions, design restraint, and quality gates. Use when writing frontend code in src/ or app/ directories.

Schema design for multi-tenant Postgres: constraints, RLS, indexes, vectors.

Applies house Supabase conventions — the new sb_publishable_/sb_secret_ API keys, RLS-first multi-tenant schema design, migration discipline, and pre-ship advisor checks. Use when creating or changing schema, auth, API keys, or client wiring on any Supabase project.

Astro + Tailwind: a sub-second page that says one thing well.

Builds a scroll-scrubbed "fly through the world" landing page where scrolling drives a pre-rendered AI camera flight through connected scenes with no visible cuts. Use when asked for a scroll cinematic, diorama landing page, 3D-world hero, or to turn a business into a scrollable world.

Hono + Postgres + OpenAPI: a typed API service with real observability.

Sessions, roles, and row-level enforcement with Supabase Auth (or Clerk).

Cron + queue + webhook glue that's idempotent and observable.

Manifest V3 extension with WXT, typed messaging, and store-ready packaging.

Retrieval-augmented chat with pgvector, hybrid search, and citations.

A fast, scriptable command-line tool with Go + Cobra (or Node + commander).

Next.js + Recharts over Postgres RPCs - fast, smooth, no jitter.

Ingest → transform → schedule with dbt, DuckDB/Postgres, and Dagster.

GitHub Actions, Docker, Terraform, and observability that pages you correctly.

Next.js storefront with Stripe Checkout and webhook-driven fulfillment.

PyTorch + Hugging Face + tracked experiments, served behind FastAPI.

One React Native codebase to both stores with Expo Router and EAS.