skill
funda-skill
Access Funda listings via a local HTTP gateway built on pyfunda
Set up
Skills are installed by placing their folder where your agent loads skills.
The download comes from github.com/risboo6909/funda-skill itself, not from our server, so what you get is exactly what its author published.
Install instructions come from the README
README
synced Oct 23, 2026
Funda Skill
Some notes for people who want to use this skill.
Disclaimer
The author is not responsible for any issues that may occur while using this skill.
Use it at your own risk.
Short Description
This is my first experiment with agent skills, so please don’t judge too harshly 🙂
At the same time, I’m very open to feedback and suggestions for improvement.
I originally built this skill for myself. OpenClaw supports running tasks on a schedule (cron), which makes it possible to do some very useful things, for example, receiving daily notifications in a messenger when new housing options appear on Funda.
Since this is an AI agent, you can describe your requirements in natural language (for example: location, price range, size, energy label, etc.). The agent can then:
- filter listings based on those criteria
- apply some basic analysis
- rank results by relevance
At some point I realized that this could also be useful for other people. Searching for housing in the Netherlands is painful, and many people rely on Funda. That’s why I decided to share this skill with the community.
This skill was developed specifically for OpenClaw. I have not tested it with other agent frameworks, so compatibility with other agents is not guaranteed.
Important Note
The initial startup of the skill may take some time.
If the agent does not respond immediately, this is expected.
On first run, the agent needs to:
- create a Python virtual environment
- install PyFunda and its Python dependencies
- start a local HTTP gateway
Security note: the gateway has no authentication or rate limiting. It should only be used in trusted local environments and should be started on loopback (127.0.0.1).
For regular updates / scheduled checks in OpenClaw / ClawHub, use Heartbeat instead of cron jobs.
Why Heartbeat is the better choice here:
- this skill depends on a local HTTP gateway process running in the same working session
- Heartbeat runs in the main session, so it can reuse that local gateway
- cron jobs may run in an isolated environment and may not be able to reach your gateway at all
- changing cron/network settings is usually not enough if the cron runtime is isolated
In practice:
- use Heartbeat for periodic notifications / monitoring
- use cron only if you fully understand your runtime/network isolation and do not depend on the local gateway session
OpenClaw Heartbeat docs:
Once the gateway (funda_gateway.py) is installed and running, subsequent requests should be much faster.
How It Works
The setup is intentionally simple.
The agent starts a local HTTP server that acts as a gateway between the agent and the Funda API.
The agent sends HTTP requests to this local server, and the server:
- calls the Funda API via PyFunda
- processes the results
- returns structured data back to the agent
The gateway is written in Python and kept intentionally minimal.
This reduces errors, makes agent interaction more reliable, and allows new features to be added later with minimal changes if needed.
Search results are returned in a consistent structured format, which makes filtering and ranking reliable for AI agents.
Final Words
If you find this skill useful, please consider giving it a star on GitHub and sharing it with friends who are also searching for housing in the Netherlands.
Thanks!