Open Reef Templates
Start from proven building blocks: production starter kits, automations, system prompts, and reusable skills.
A curated prompt pattern set for role framing, constraints, output formats, and tool invocation style. It is a fast way to raise baseline prompt quality across teams.
Cloudflare’s agents starter is optimized for low-latency, globally distributed agent workloads. It combines edge runtime patterns with Workers AI so you can prototype and deploy agent endpoints without managing traditional servers.
This workflow template ingests documents, indexes embeddings, and serves grounded answers through retrieval-augmented generation. It is a common production pattern for internal knowledge assistants.
This starter kit gives teams a complete local-first AI stack with very little setup friction. You get n8n for orchestration, Ollama for private model serving, Qdrant for vector search, and PostgreSQL for durable workflow state. It is one of the fastest ways to stand up secure agent automation on your own infrastructure.
A curated path for installing and combining coding-focused OpenClaw skills from the ecosystem. It reduces setup time for teams that want a capable software engineering pod quickly.
The OpenClaw starter kit packages practical conventions for memory files, role definitions, and agent coordination loops. It is designed for teams that want persistent state and clean delegation boundaries between strategy, engineering, and research agents.
TheBigPromptLibrary is a broad repository covering system prompts, custom instructions, and security-adjacent examples. It is useful as a comparative corpus when refining internal prompt standards.
Generates comprehensive API reference documentation from code, OpenAPI specs, or endpoint descriptions. Creates clear endpoint descriptions, parameter tables, request/response examples, and error handling guidance that developers can actually use.
A comprehensive system prompt that transforms an AI agent into an executive assistant capable of managing calendars, prioritizing communications, drafting responses, and handling high-level administrative tasks with appropriate discretion and business acumen.
Blueprints focused on instruction hierarchy, safety boundaries, and nuanced writing behavior for Claude-style models. They help teams produce consistent assistant behavior across tasks.
A detailed prompt for AI-powered code review that checks for bugs, security vulnerabilities, performance issues, and adherence to coding standards. Provides constructive feedback with specific suggestions and catches issues human reviewers often miss.
This collection focuses on coding assistant behavior: task decomposition, verification habits, and concise patch explanations. It is highly relevant for engineering teams deploying code agents.
Automated workflow that monitors competitor websites, social media, and news sources for significant changes, product launches, pricing updates, or strategic moves. Summarizes findings and alerts stakeholders when action may be needed.
End-to-end workflow for content creation from idea generation through scheduling and publication. Generates content ideas, creates drafts based on templates and brand voice, optimizes for SEO, and schedules publication across multiple channels.
Automated workflow that enriches CRM leads with additional data, scores them based on fit and intent signals, segments them into priority tiers, and routes high-value leads to sales while nurturing others through automated sequences.
A customer-facing prompt that creates an empathetic, knowledgeable support agent capable of troubleshooting issues, explaining product features, gathering feedback, and escalating complex cases appropriately while maintaining a helpful, professional tone.
An analytical prompt that enables an AI agent to perform exploratory data analysis, identify trends and anomalies, create visualizations, and generate insights from datasets with clear explanations of statistical methods and confidence levels.
A starter template for creating AI-powered Discord bots. Includes slash command handling, message listeners, voice channel integration, role-based permissions, and OpenAI integration for intelligent responses with conversation memory.
Workflow that helps job seekers manage their application pipeline. Tracks applications across multiple job boards, monitors status changes, sets follow-up reminders, generates customized resumes and cover letters, and provides interview preparation materials.
LangGraph offers a robust model for stateful agent flows, retries, and branching logic. This template emphasizes graph-based orchestration where each step is explicit, inspectable, and easier to test than monolithic prompt chains.
A careful, detail-oriented prompt for reviewing legal documents, contracts, and agreements. Identifies potential issues, ambiguous language, missing clauses, and unfavorable terms while maintaining appropriate disclaimers about not providing legal advice.
Aggregates content from multiple RSS feeds, filters based on keywords and relevance criteria, summarizes articles into digestible snippets, removes duplicates across sources, and delivers a curated daily or weekly digest via email or messaging apps.
Designed for multilingual customer support agents that must preserve tone while adapting for locale. The template includes fallback behavior when confidence is low.
A beginner-friendly workflow that demonstrates the full request-to-response loop in n8n. It shows how to collect user input, generate model output, and structure replies in a maintainable visual graph.
Generates first-draft support responses based on issue type, account metadata, and knowledge snippets. The workflow speeds up queue handling while preserving a review step for sensitive cases.
Automate inbox triage by classifying incoming emails, extracting intent, and routing actions to the right systems. This reduces repetitive ops work and gives teams faster response cycles.
Monitors pull requests and generates review summaries based on changed files, labels, and policy checks. It helps engineering teams surface risk early in the review cycle.
Extracts key fields from invoices and maps them into accounting or ERP systems. This template reduces manual entry errors and speeds up reconciliation workflows.
This template enriches inbound leads with company context, role data, and prioritization signals. It is useful for sales ops teams who want better handoff quality between inbound capture and outreach.
Transforms raw meeting notes into structured tasks with owners, due dates, and next actions. The core benefit is reducing the gap between discussion and execution.
Collects standup updates from Slack channels and produces concise daily summaries for leads. This template helps teams preserve context without forcing everyone to read long threads.
Converts long-form content into platform-specific drafts for social channels. The workflow accelerates distribution while maintaining brand voice constraints.
This template is a practical baseline for shipping chat-style and assistant-style experiences with Next.js. It includes modern app routing conventions and a developer-friendly structure for iterating quickly on prompts, UI states, and API handlers.
This skill grouping targets browser control workflows for QA, scraping, and repetitive web tasks. It enables robust UI interactions with snapshots, selectors, and scripted actions.
Focused on deployment, runtime operations, and cloud workflows, this pack helps agents participate in infrastructure tasks under supervision. It can streamline repetitive SRE operations.
PromptOps skills create a repeatable loop for drafting prompts, running evaluations, and tracking quality over time. This turns prompt work from ad hoc craft into an engineering process.
A practical set of research-oriented skills for web search, source synthesis, and evidence-backed summaries. It supports faster decision-making with stronger citation hygiene.
Workflow for extracting structured information from PDF documents like invoices, contracts, resumes, or reports. Uses OCR and parsing to pull key fields into databases or spreadsheets, enabling search, analysis, and automation on previously locked-in-PDF data.
A strategic prompt that enables an AI agent to assist with product management tasks including user story writing, feature prioritization, competitive analysis, roadmap planning, and stakeholder communication with appropriate product thinking frameworks.
A complete Retrieval-Augmented Generation starter kit using local embedding models and vector databases. Enables semantic search over private documents, integrates with LLMs for grounded responses, and runs entirely on your infrastructure.
A specialized subset of prompts centered on refusal boundaries, escalation pathways, and risk-sensitive tool usage. It helps teams reduce policy drift across complex agent systems.
A specialized prompt for generating optimized SQL queries from natural language requests. Produces correct, efficient queries with proper joins, filters, and aggregations while explaining the query logic and suggesting performance improvements.
A system prompt that configures an AI agent to produce clear, accurate technical documentation including API docs, user guides, tutorials, and architecture documents. Emphasizes clarity, accuracy, and appropriate detail level for different audiences.
A minimal starter template for building a Telegram bot powered by OpenAI's API. Handles message routing, maintains conversation context per user, includes command handlers, and provides a foundation for adding custom bot capabilities.
A targeted set of prompts for UI generation and component scaffolding. It provides framing for style constraints, accessibility, and predictable output structure.
End-to-end voice interaction starter kit that connects speech-to-text, language model processing, and text-to-speech into a conversational voice assistant. Handles audio input/output, conversation state, and real-time response streaming.
Automated workflow that pulls metrics from multiple sources, formats them into executive-ready reports with visualizations and narrative summaries, highlights significant changes or anomalies, and delivers via email or Slack on a recurring schedule.
