Cached Google/DDG-Search to reduce rate limit penalty #2338

[janpdevops]

Cached Google/DDG-Search to reduce rate limit penalty Issue #2338

See #2338

• Summary of changes: I added a cached version of the Google Search. Also, an example on how to use it is included in cookbook/agent_concepts/other/google_cache_demo.py
• Related issues: Cached Google/DDG-Search to reduce rate limit penalty #2338
• Motivation and context:When developing, you will pretty soon hit a rate limit with DDG and, later with google. Since the requests are repetitive, you can avoid this with a cache.
• Environment or dependencies: You will need a database as in the pgvector example.
• Impact on metrics: (If applicable) Describe changes in any metrics or performance benchmarks.

Fixes # (2338)

---

Type of change

Please check the options that are relevant:

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Model update (Addition or modification of models)
• Other (please describe):

---

Checklist

• Adherence to standards: Code complies with Agno’s style guidelines and best practices.
• [ ]
• Formatting and validation: You have run ./scripts/format.sh and ./scripts/validate.sh to ensure code is formatted and linted. No, please add the .bat files again.
• Self-review completed: A thorough review has been performed by the contributor(s).
• Documentation: Docstrings and comments have been added or updated for any complex logic.
• Examples and guides: Relevant cookbook examples have been included or updated (if applicable).
• Tested in a clean environment: Changes have been tested in a clean environment to confirm expected behavior.
• Tests (optional): Tests have been added or updated to cover any new or changed functionality.

---

Additional Notes

I am still new to python, so I am open to suggestions. I am willing to help the project along.

[dirkbrnd]

Love the contribution. We think it should be done a little differently, i'd rather have an in-memory cache managed by a decorator, which we can easily add to more functions.
We just haven't gotten around to it.

[janpdevops]

@dirkbrnd
I think I could create a wrapper. That would be used in the Agent similar like this:
tools=[CachedSearchTool(
wrappedTool=GoogleSearchTools(
fixed_max_results=10,
fixed_language="en"
),
),
db_url="postgresql://ai:ai@localhost:5532",
table_name="search_cache",
)],

You could use Duck Duck Go or whatever you like instead of google.

[janpdevops]

Hi, I changed the cache to a wrapper of another search tool. It worked with the Duck Duck Go and Google.

[janpdevops]

Hi @dirkbrnd,
I choose the database instead of a in memory cache - this is useful, when you run the small tutorial examples (that just perform a couple of steps and terminate). It was useful for me, when I had to run the same code again and again to experiment. If you think this is a step, I'd be happy to merge.

[dirkbrnd]

Hi @janpdevops
I have since introduced caching on some of our tools (I can easily add to more). Check out the updates to DDG tools for example.
If that works for you, I suggest we close this PR. I'll do so in the mean time, but please let me know if that is not working.