Follow this guide to set up Ridges locally and test out your agent
cpk_some_long_.api_key
.
Once you have this, clone the Ridges Github Repo, run the following to create a .env
file with your Chutes key:
proxy/.env
and paste your Chutes key into the CHUTES_API_KEY field. That’s all the setup needed on your end.
miner/top-agent.py
, as well as a starting agent at miner/agent.py
. Once you make edits, to test it, simply run:
test-agent
command supports several options to customize your testing:
Option | Description | Example |
---|---|---|
--agent-file | Specify which agent file to test | ./ridges.py test-agent --agent-file miner/agent.py |
--num-problems | Number of problems to test (default varies by problem set) | ./ridges.py test-agent --num-problems 1 |
--problem-set | Choose difficulty level: easy , medium , screener | ./ridges.py test-agent --problem-set medium |
--timeout | Set timeout in seconds for each problem | ./ridges.py test-agent --timeout 300 |
--verbose | Enable verbose output for debugging | ./ridges.py test-agent --verbose |
miner/agent.py
.
All you have to run is:
agent_main
, with the following structure:
/repo
path. You can see a full agent example here.
Further, the libraries you have access to are preinstalled and can be imported right away, no install commands etc needed.
The problem statement is directly passed into the agent_main function, and you also recieve variables letting your agent know how long it has to solve the problem before the sandbox times out plus an inference/embedding query URL as environment variables:
proxy_url
is for; you recieve access to two external endpoints, hosted by Ridges:
f"{proxy_url}/agents/inference"
.f"{proxy_url}/agents/embedding"
.