Pydantic Chat
A local-first chat UI for your Pydantic AI agents.
uvx pydantic-work <module>:<agent variable>
# e.g. uvx pydantic-work chat.my_agent:agent
What is it?
Pydantic Chat provides a beautiful web interface for interacting with your Pydantic AI agents. Your agent runs locally on your machine, and the chat UI can be accessed either via localhost or through a secure remote URL.
Important: your messages never leave your machine and never touch the internet. Remote URLs are a convenience for you to run multiple agents on your machine without picking and keeping track of ports. Anyone visiting one of your URLs will get an empty chat window and a message like below ↓
Installation
# Run with uvx
uvx pydantic-work
# Or clone the repo and run locally
gh repo clone dsfaccini/ai-chat-ui
cd agent
uv run pydantic-work
Usage
# Or install with uv as a tool
uv tool install pydantic-work
# Basic usage
uvx pydantic-work module.path:agent_variable
# Example
pydantic-work chat.golden_gate_bridge:agent
# Localhost-only mode (skip remote registration)
pydantic-work --localhost chat.golden_gate_bridge:agent
# Custom port
pydantic-work --port 8000 chat.golden_gate_bridge:agent
How It Works
- Local Server: Your agent runs on your machine with a FastAPI server
- Remote Access (Optional): On first run, you'll be prompted to choose a project slug (e.g.,
my-project) - Two URLs: Access your chat via:
- Local:
http://127.0.0.1:PORT - Remote:
https://your-project.pydantic.work/(if registered)
- Local:
Your agent code and data never leave your machine. The remote URL just provides the frontend.
Note: The localhost UI is served via CDN (jsdelivr) at a pinned version, while the remote UI is served from the Cloudflare Worker and may be on a different version. Both UIs are compatible with the same backend API.
Example Agent
# src/my_agent.py
from pydantic_ai import Agent
agent = Agent(
'anthropic:claude-sonnet-4-0',
instructions="You are a helpful assistant."
)
@agent.tool_plain
def get_current_time() -> str:
"""Get the current time."""
from datetime import datetime
return datetime.now().strftime("%I:%M %p")
Run it:
pydantic-work chat.my_agent:agent
Configuration
On first run, Pydantic Chat creates .pydantic-work/config.json in your project directory with your slug, token, and port. This folder is automatically added to .gitignore.
Patterns
Multiple Projects
Each project directory gets its own config, so you can run multiple agents with different slugs:
cd project-a && pydantic-work agent:agent # -> project-a.pydantic.work
cd project-b && pydantic-work agent:agent # -> project-b.pydantic.work
Troubleshooting
Failed to load agent: Check that your module path is correct (module.path:variable_name).
Registration failed: The server will automatically fall back to localhost mode. Use --localhost to skip registration entirely.
Slug already taken: Choose a different slug when prompted. Slugs are globally unique.
Links
Metadata
Release files for pydantic-work 0.1.11
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| pydantic_work-0.1.11.tar.gz | 15.0 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| pydantic_work-0.1.11-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 30.9 kB
Release files / pydantic_work-0.1.11.tar.gz
| Download URL | pydantic_work-0.1.11.tar.gz |
|---|---|
| Size | 15.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f790f1171a199a15ddc06c509f6386582e18c281596774b0aaf5f0105229b3c7
|
|
BLAKE2b-256 checksum How to use checksums |
5993d183a4e62fb180a6732d1ff14d18e3abfe98f6f7634526bdeb6a64aa98fe
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.7.12
|
Release files / pydantic_work-0.1.11-py3-none-any.whl
| Download URL | pydantic_work-0.1.11-py3-none-any.whl |
|---|---|
| Size | 15.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
65d3873e1b15e7aa552d5b3d6a13d6107e524f7296ee282bff6bddae10b1a344
|
|
BLAKE2b-256 checksum How to use checksums |
8725a195c52729fc4d4bc7bfec2fe458bbbc86a433ff04308e57a19236f15b2c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.7.12
|
Security for pydantic-work 0.1.11
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.