
Prompts Chat
github.com/f/prompts.chat- Category
- AI Tools
- Rank
- No. 26Tools index
- Pricing
- Open Source
- Type
- APP
- Builder
- f
- GitHub
- 169.6k stars
- Date
About
The world's largest open-source library of AI prompts for ChatGPT, Claude, Gemini, and other AI models. Browse thousands of curated prompts, contribute your own, and self-host for your organization with complete privacy.
What it does
Prompts Chat combines a searchable prompt collection with tools for running a branded internal library. People can browse, submit, categorize, and share prompts, while teams can configure authentication, private prompts, themes, languages, and optional AI features. Prompts also reach other tools through command-line, plugin, dataset, raw document, and MCP interfaces.
Why it's ranked here
Its strongest case is breadth beyond a static prompt list. The same project supports public discovery, organizational deployment, reusable data, education, and direct tool integration. The configuration surface is unusually practical for teams. That breadth brings meaningful operational weight, including a database, authentication, and a substantial web stack.
What's good
The self-hosted edition exposes concrete controls for branding, color, layout density, registration, authentication providers, languages, and individual features. PostgreSQL persistence supports a real shared collection rather than a folder of snippets. Prompts remain portable through CSV, Markdown-style documents, a hosted dataset, raw responses, and MCP. Automated tests and coverage commands are configured.
Tradeoffs
Self-hosting is not lightweight. It expects Node.js 24, PostgreSQL, dependency installation, schema generation, migrations, and seeding. Several optional capabilities require additional service configuration. Production browser monitoring enables session replay sampling, logging, trace sampling, and default personal-data transmission. Organizations promising strict privacy should audit and adjust that monitoring configuration before deployment.
How to use it well
Use it when a team needs a governed, searchable prompt commons that also feeds assistants and command-line workflows. Start with browsing and adapting existing prompts, then self-host when branding, access control, or private material matters. It teaches prompting, but it does not replace model hosting, evaluation infrastructure, or application-specific AI orchestration.
Technical notes+
The root package.json defines a private Next.js 16 and React 19 application, requires Node 24.x, uses Prisma with PostgreSQL, and provides Vitest test and coverage scripts. packages/prompts.chat/tsup.config.ts builds the TypeScript library as CommonJS and ESM with declarations and source maps, plus an ESM CLI with a Node shebang. src/proxy.ts rewrites .prompt.md and .prompt.yml URLs to raw API responses. src/lib/config/index.ts applies PCHAT_ runtime overrides for branding, themes, authentication, locales, and feature flags. src/lib/auth/index.ts builds NextAuth providers dynamically over a Prisma adapter. src/instrumentation-client.ts enables production Sentry tracing, logs, replay sampling, error replay, and sendDefaultPii.
Observed
- License
- MIT for source code and site-authored content; CC0 1.0 Universal for prompt content and data
- Primary language
- TypeScript, with Next.js and React for the web application
- Install surface
- Runs through npx for setup, CLI use, and the local MCP server; manual setup uses npm
- Interfaces
- Website, CLI, Claude Code plugin, remote or local MCP server, raw prompt URLs, CSV, and hosted dataset
- Runtime and storage
- Node.js 24.x and PostgreSQL through Prisma
- Library packaging
- CommonJS and ESM builds with TypeScript declarations and source maps
Read from README.md, package.json, src/proxy.ts, src/instrumentation.ts, src/instrumentation-client.ts, src/app/error.tsx, src/app/global-error.tsx, packages/prompts.chat/tsup.config.ts, packages/raycast-extension/raycast-env.d.ts, packages/raycast-extension/eslint.config.mjs, src/lib/auth/index.ts, src/lib/i18n/index.ts, src/lib/i18n/server.ts, src/lib/config/index.ts, src/lib/plugins/index.ts.
What it can do
Browse thousands of curated AI prompts
Search query or category selection → List of relevant AI prompts for ChatGPT, Claude, Gemini, and other models
Contribute and submit custom AI prompts
User-created prompt text and metadata → Published prompt in the open-source library
Self-host prompt library for organizations
Organization's infrastructure and deployment configuration → Private instance of the prompt library with complete data privacy
Search and filter prompts by AI model compatibility
AI model type (ChatGPT, Claude, Gemini, etc.) and search criteria → Filtered list of compatible prompts for the specified AI model
Access prompt examples for specific use cases
Use case category or task type → Curated prompt templates ready for use with AI chat models
Tags
Tech Stack
Media

Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.