PostHog / PostHog/wizard

09 — Decide, ramp, and decouple

Open
#529 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
TypeScript
Stars
197
Forks
51
Avg merge
1d 23h
Merged PRs (30d)
94

Description

Epic: #520 · Behavior change: ramp · Depends on: #527, #528

Summary

Use the metrics + parity scorecard to decide whether pi replaces anthropic, then ramp it — and if so,
drop the coupled SDK.

Scope

  • Review #527 dashboards + #528 scorecard; decide pi vs anthropic.
  • Ramp the winner via the existing wizard-runner 25 → 50 → 100 (each step gated on metrics); revert to
    anthropic stays one flip.
  • If pi wins: flip the default runner in DEFAULT_ROUTE/ROUTES to pi, remove
    @anthropic-ai/claude-agent-sdk + the anthropic runner from RUNNERS, and clean up the second SDK
    consumer (mcp-prompt-streaming.ts).
  • If anthropic still wins: keep it the default runner, leave the seam + pi runner registered behind the
    flag for re-evaluation.

Acceptance criteria

  • Decision recorded with the data behind it.
  • Winner ramped to target, or status quo retained with rationale.
  • If decoupling: dep removed, no source imports it, all programs/frameworks green on the new default.

Contributor guide

No contributing guide indexed for this repository

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

Start by reviewing #527's dashboards and #528's scorecard, then inspect the existing wizard-runner ramp and the DEFAULT_ROUTE, ROUTES, and RUNNERS entry points. If pi wins, check the @anthropic-ai/claude-agent-sdk dependency, the anthropic runner, and mcp-prompt-streaming.ts for cleanup. Done means the decision and data are recorded, the winner reaches its target or the status quo is justified, and the acceptance checks pass.

Written by the indexing model from the issue text.

Assessment

Tech stack
typescript
Domain
tooling
Issue type
Refactor
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
38/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.