Build reliable customer-facing AI agents with Parlant: an interaction control harness optimized for controlled, consistent, and predictable LLM interactions.
How to install parlant?
emcie-co/parlant
$git clone https://github.com/emcie-co/parlantInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Install parlant by running `git clone https://github.com/emcie-co/parlant`, then use it for the current task and follow its documentation at https://github.com/emcie-co/parlant.
README.md
| 1 | <div align="center"> |
| 2 | |
| 3 | <picture> |
| 4 | <source media="(prefers-color-scheme: dark)" srcset="https://github.com/emcie-co/parlant/blob/develop/docs/LogoTransparentLight.png?raw=true"> |
| 5 | <img alt="Parlant - AI Agent Framework" src="https://github.com/emcie-co/parlant/blob/develop/docs/LogoTransparentDark.png?raw=true" width=400 /> |
| 6 | </picture> |
| 7 | |
| 8 | <h3>Finally, LLM agents that actually follow instructions</h3> |
| 9 | |
| 10 | <p> |
| 11 | <a href="https://www.parlant.io/" target="_blank">🌐 Website</a> • |
| 12 | <a href="https://www.parlant.io/docs/quickstart/installation" target="_blank">⚡ Quick Start</a> • |
| 13 | <a href="https://discord.gg/duxWqxKk6J" target="_blank">💬 Discord</a> • |
| 14 | <a href="https://www.parlant.io/docs/quickstart/examples" target="_blank">📖 Examples</a> |
| 15 | </p> |
| 16 | |
| 17 | <p> |
| 18 | <!-- Keep these links. Translations will automatically update with the README. --> |
| 19 | <a href="https://zdoc.app/de/emcie-co/parlant">Deutsch</a> | |
| 20 | <a href="https://zdoc.app/es/emcie-co/parlant">Español</a> | |
| 21 | <a href="https://zdoc.app/fr/emcie-co/parlant">français</a> | |
| 22 | <a href="https://zdoc.app/ja/emcie-co/parlant">日本語</a> | |
| 23 | <a href="https://zdoc.app/ko/emcie-co/parlant">한국어</a> | |
| 24 | <a href="https://zdoc.app/pt/emcie-co/parlant">Português</a> | |
| 25 | <a href="https://zdoc.app/ru/emcie-co/parlant">Русский</a> | |
| 26 | <a href="https://zdoc.app/zh/emcie-co/parlant">中文</a> |
| 27 | </p> |
| 28 | |
| 29 | <p> |
| 30 | <a href="https://pypi.org/project/parlant/"><img alt="PyPI" src="https://img.shields.io/pypi/v/parlant?color=blue"></a> |
| 31 | <img alt="Python 3.10+" src="https://img.shields.io/badge/python-3.10+-blue"> |
| 32 | <a href="https://opensource.org/licenses/Apache-2.0"><img alt="License" src="https://img.shields.io/badge/license-Apache%202.0-green"></a> |
| 33 | <a href="https://discord.gg/duxWqxKk6J"><img alt="Discord" src="https://img.shields.io/discord/1312378700993663007?color=7289da&logo=discord&logoColor=white"></a> |
| 34 | <img alt="GitHub Repo stars" src="https://img.shields.io/github/stars/emcie-co/parlant?style=social"> |
| 35 | </p> |
| 36 | |
| 37 | <a href="https://trendshift.io/repositories/12768" target="_blank"> |
| 38 | <img src="https://trendshift.io/api/badge/repositories/12768" alt="Trending on TrendShift" style="width: 250px; height: 55px;" width="250" height="55"/> |
| 39 | </a> |
| 40 | |
| 41 | </div> |
| 42 | |
| 43 | ## 🎯 The Problem Every AI Developer Faces |
| 44 | |
| 45 | You build an AI agent. It works great in testing. Then real users start talking to it and... |
| 46 | |
| 47 | - ❌ It ignores your carefully crafted system prompts |
| 48 | - ❌ It hallucinates responses in critical moments |
| 49 | - ❌ It can't handle edge cases consistently |
| 50 | - ❌ Each conversation feels like a roll of the dice |
| 51 | |
| 52 | **Sound familiar?** You're not alone. This is the #1 pain point for developers building production AI agents. |
| 53 | |
| 54 | ## ⚡ The Solution: Stop Fighting Prompts, Teach Principles |
| 55 | |
| 56 | Parlant flips the script on AI agent development. Instead of hoping your LLM will follow instructions, **Parlant ensures it**. |
| 57 | |
| 58 | ```python |
| 59 | # Traditional approach: Cross your fingers 🤞 |
| 60 | system_prompt = "You are a helpful assistant. Please follow these 47 rules..." |
| 61 | |
| 62 | # Parlant approach: Ensured compliance ✅ |
| 63 | await agent.create_guideline( |
| 64 | condition="Customer asks about refunds", |
| 65 | action="Check order status first to see if eligible", |
| 66 | tools=[check_order_status], |
| 67 | ) |
| 68 | ``` |
| 69 | |
| 70 | - ✅ [Blog: How Parlant Ensures Agent Compliance](https://www.parlant.io/blog/how-parlant-guarantees-compliance) |
| 71 | - 🆚 [Blog: Parlant vs LangGraph](https://www.parlant.io/blog/parlant-vs-langgraph) |
| 72 | - 🆚 [Blog: Parlant vs DSPy](https://www.parlant.io/blog/parlant-vs-dspy) |
| 73 | - ⚙️ [Blog: Inside Parlant's Guideline Matching Engine](https://www.parlant.io/blog/inside-parlant-guideline-matching-engine) |
| 74 | |
| 75 | #### Parlant gives you all the structure you need to build customer-facing agents that behave exactly as your business requires: |
| 76 | |
| 77 | - **[Journeys](https://parlant.io/docs/concepts/customization/journeys)**: |
| 78 | Define clear customer journeys and how your agent should respond at each step. |
| 79 | |
| 80 | - **[Behavioral Guidelines](https://parlant.io/docs/concepts/customization/guidelines)**: |
| 81 | Easily craft agent behavior; Parlant will mat |
Related
6 picksType
- langchain-mcp-adaptersMake Anthropic Model Context Protocol (MCP) tools compatible with LangChain and LangGraph agents.Jul 2026
- mcporterTypeScript runtime and CLI for connecting to configured Model Context Protocol servers.Jul 2026
- mcp-proxyA MCP server which proxies requests to a remote MCP server over streamable HTTP or SSE.Jul 2026
- mcp-atlassianThe Model Context Protocol (MCP) Atlassian integration is an open-source implementation that bridges Atlassian products (Jira and Confluence) with AI language models following Anthropic's MCP specification.Jul 2026
- open-webuiOpen WebUIJul 2026
- rulesyncUnified AI rules management CLI tool that generates configuration files for various AI development toolsJul 2026
