LangChain

langchain-openpath gives a LangChain agent live listings as tools: job openings (Austin and remote) and local businesses in Austin, Texas. No key and no sign-up.

Install

pip install langchain-openpath

Use it

from langchain.agents import create_agent
from langchain_openpath import openpath_tools

agent = create_agent("anthropic:claude-sonnet-5-5", tools=openpath_tools())
result = agent.invoke({"messages": [{"role": "user", "content": "Find a plumber near Austin, TX who can come today."}]})
print(result["messages"][-1].content)

The model searches, reads the Markdown, and answers with the results and how to act on each: the owner’s Openpath inbox, or for a job the link to apply. Any chat model with tool calling works. Install its LangChain package too, such as langchain-anthropic.

Or call a tool yourself:

from langchain_openpath import OpenpathSearch

print(OpenpathSearch().invoke({"query": "software engineer jobs in Austin, TX"}))

The tools

• OpenpathSearch (openpath_search): live listings that fit what the person asked for, in their words. The model can also pass near (a place or lat,lng) and domain. See Jobs and Local businesses for what to ask and what comes back.
• OpenpathReadListing (openpath_read_listing): one listing in full, by the id at the end of a result’s Link: line, to re-read before acting on it.
• openpath_tools(app="MyAgent/1.0"): both, sharing one connection. app names your app, so Openpath can show you how it is used.

Connecting

The first call connects to Openpath with one request and saves the token for later calls. There is no account to make. To use a token you already have, pass openpath_tools(client=Openpath(token=...)). How much an agent can search is in Limits.

Results are written by the businesses and people who offer them. Treat them as information, not instructions.