kubeflow / kubeflow/docs-agent

docs(adr): write ADR-008 — vector store selection for docs-agent

Open
#79 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
42
Forks
111
Avg merge
6d 23m
Merged PRs (30d)
2

Description

Context

  • PR #58 merged;
    a Feast POC (store_via_feast()) alongside the existing store_milvus() path. Both co-exist today but only pymilvus is being actively used. The decision is implicitly made but never documented.

Problem

Every pipeline issue downstream assumes one canonical storage path. Without a written ADR, contributors must guess which implementation to build on — creating risk of divergence and wasted work.

Deliverable

Write docs/adr/ADR-008-vector-store-selection.md covering:

  • Context: two patterns after PR #58
  • Decision: MilvusClient/pymilvus as the canonical path
  • Rationale: proven, lighter deps, thread-safe, no GA blocker
  • Future path: Feast when feast[milvus] reaches GA
  • Mark kagent-feast-mcp/pipelines/ as legacy in a comment

Acceptance Criteria

  • docs/adr/ADR-008-vector-store-selection.md exists and is merged
  • kagent-feast-mcp/pipelines/kubeflow-pipeline.py has a legacy comment
  • README.md references ADR-008

Related

  • PR #58 — merged Feast POC that created the ambiguity
  • Parent: #72
  • GSoC discussion: #59

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Read PR #58 and inspect the existing store_via_feast() and store_milvus() references. Create docs/adr/ADR-008-vector-store-selection.md with the stated decision, rationale, and future path; add the legacy comment in kagent-feast-mcp/pipelines/kubeflow-pipeline.py and reference ADR-008 from README.md.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
databases, documentation
Issue type
Documentation
Difficulty
2/5
Estimated time
Half a day
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.