# PromptRails

> PromptRails is an AI agent orchestration platform for building, deploying, and monitoring LLM-powered applications. It provides prompt management, agent orchestration, distributed tracing, guardrails, and multi-provider LLM support.

- Multi-provider support: OpenAI, Anthropic, Google Gemini, DeepSeek, Fireworks, xAI, OpenRouter
- Agent types: simple, chain, multi-agent, workflow, composite
- 14 built-in guardrail scanners for safety and compliance
- Python and JavaScript/TypeScript SDKs, CLI, MCP server
- Data source integrations: PostgreSQL, MySQL, MSSQL, BigQuery, Snowflake, Redshift, ClickHouse

## Docs

- [Introduction](https://docs.promptrails.ai/introduction.md): Overview of PromptRails platform and key features
- [Quickstart](https://docs.promptrails.ai/quickstart.md): Get up and running in under 5 minutes
- [Agents](https://docs.promptrails.ai/agents.md): Build agents using five execution strategies
- [Agent Versioning](https://docs.promptrails.ai/agent-versioning.md): Version control and promotion for agents
- [Prompts](https://docs.promptrails.ai/prompts.md): Versioned prompts with Jinja2 templating and schemas
- [Prompt Versioning](https://docs.promptrails.ai/prompt-versioning.md): Version control and promotion for prompts
- [LLM Gateway](https://docs.promptrails.ai/llm-gateway.md): Use hosted models without managing provider keys
- [Data Sources](https://docs.promptrails.ai/data-sources.md): Query databases directly from agents
- [Knowledge Sources](https://docs.promptrails.ai/knowledge-sources.md): Index PDFs, pasted notes, and GitHub Markdown/MDX for retrieval
- [Credentials](https://docs.promptrails.ai/credentials.md): Manage credentials for LLM providers and databases
- [MCP Tools](https://docs.promptrails.ai/mcp-tools.md): Model Context Protocol tool integrations
- [Guardrails](https://docs.promptrails.ai/guardrails.md): Input and output safety scanners
- [Executions](https://docs.promptrails.ai/executions.md): Execution records and status tracking
- [Tracing](https://docs.promptrails.ai/tracing.md): Distributed tracing with span hierarchy
- [Sessions & Chat](https://docs.promptrails.ai/sessions-and-chat.md): Multi-turn conversation management
- [Cost Tracking](https://docs.promptrails.ai/cost-tracking.md): Per-execution cost calculation across providers
- [API Keys & Scopes](https://docs.promptrails.ai/api-keys-and-scopes.md): Fine-grained API key permissions
- [Webhook Triggers](https://docs.promptrails.ai/webhook-triggers.md): Event-driven webhook notifications
- [Apps](https://docs.promptrails.ai/apps.md): Build hosted apps, dashboards, forms, portals, and presentation decks
- [Approvals](https://docs.promptrails.ai/approvals.md): Human-in-the-loop approval workflows
- [Assets and Media](https://docs.promptrails.ai/assets.md): Use media tools inside agents and manage generated files
- [Python SDK](https://docs.promptrails.ai/python-sdk.md): Official Python client library
- [JavaScript SDK](https://docs.promptrails.ai/javascript-sdk.md): Official JavaScript/TypeScript client library
- [Go SDK](https://docs.promptrails.ai/go-sdk.md): Official Go client library
- [CLI](https://docs.promptrails.ai/cli.md): Command-line interface for PromptRails
- [MCP Server](https://docs.promptrails.ai/mcp-server.md): MCP server for IDE integration
- [A2A Protocol](https://docs.promptrails.ai/a2a-protocol.md): Agent-to-Agent communication protocol
- [Local Emulator](https://docs.promptrails.ai/local-emulator.md): Run PromptRails flows locally for development and testing
- [Skills](https://docs.promptrails.ai/skills.md): Define reusable agent capabilities and instructions
- [Desktop Monitor](https://docs.promptrails.ai/desktop-monitor.md): Inspect desktop agent sessions and activity
- [API Reference](https://docs.promptrails.ai/rest-api-reference.md): REST API conventions and authentication

## Optional

- [Workspace Management](https://docs.promptrails.ai/workspace-management.md): Create and manage workspaces
- [Team & Roles](https://docs.promptrails.ai/team-and-roles.md): Role-based access control
- [Billing & Plans](https://docs.promptrails.ai/billing-and-plans.md): Plans, limits, and billing
- [Security](https://docs.promptrails.ai/security.md): Security practices and best practices

## Full content

- [llms-full.txt](https://docs.promptrails.ai/llms-full.txt): Consolidated dump of the full documentation corpus — one fetch instead of ~25 separate crawls.