Split GCP `laws` RAG into separate jurisdictions
@yangm2 is already working on this.
Since Aug 31, 2026.
Assessment
This issue has not been assessed yet.
Description
Currently we have a monolithic GCP thingy (cloud object bucket -> data store -> search app) that encapsulates laws for
- Oregon state
- Portland
- Eugene
(as documented here)
I propose that since the LLM can call more than 1 tool (including more than 1 RAG) and can call them in parallel, it would be more scalable to split the jurisdictions into separate RAGs.
Pros:
- update/maintain each jurisdiction independently (updating one or more should still require an app deployment, but is less disruptive than reindexing unchanged jurisdictions)
- incrementally add/remove jurisdictions (e.g. counties? Washington state? Vancouver, WA? Bend, OR? Federal?)
- simplify the RAG call (remove the city/state arguments ... those would be encoded in the tool name/docstring)
- each RAG could be configured independently since Oregon's laws are many times larger than Portland's or Eugene's laws, the same settings (e.g. max_documents) don't really make sense
- evaluating and debugging traces in LangSmith could be simplified by just looking at whether the right tools were called or not
Cons:
- increases the number of thingys (buckets -> data stores -> search apps) that need to be configured and maintained
- if parallel tool calls aren't well supported, the responses might slow down due to serially calling the various RAGs
- Dominant language
- Python
- Stars
- 13
- Forks
- 21
- Avg merge
- 6d 7h
- Merged PRs (30d)
- 3
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from codeforpdx/tenantfirstaid
-
2026 roadmap Open
Difficulty 5/5 Over a week Newbie friendliness 25/100
codeforpdx/tenantfirstaid#390 ·
-
backend good first issue infrastructure
Difficulty 3/5 1-2 days Newbie friendliness 48/100
codeforpdx/tenantfirstaid#389 ·
-
AI Prompts backend external-RAG infrastructure
Difficulty 4/5 3-5 days Newbie friendliness 45/100
codeforpdx/tenantfirstaid#372 · 2 comments ·
-
bug frontend
Difficulty 4/5 3-5 days Newbie friendliness 38/100
codeforpdx/tenantfirstaid#371 ·
-
Difficulty 4/5 3-5 days Newbie friendliness 48/100
codeforpdx/tenantfirstaid#355 ·
All issues in codeforpdx/tenantfirstaid
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
bancolombia/sentinel#23 ·
-
test md OpenCI
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
-
integration:quickjs org:external priority:backlog topic:code-interpreter topic:middleware type:feature
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
langchain-ai/deepagents#6450 ·
-
bug client
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 74/100