Skip to content

fix(wizard): Whack a mole part 5: report per-task instead of the unused binding model on orchestrator runs - #1019

Open
gewenyu99 wants to merge 5 commits into
mainfrom
posthog-code/whack-a-mole-5
Open

fix(wizard): Whack a mole part 5: report per-task instead of the unused binding model on orchestrator runs#1019
gewenyu99 wants to merge 5 commits into
mainfrom
posthog-code/whack-a-mole-5

Conversation

@gewenyu99

@gewenyu99 gewenyu99 commented Jul 28, 2026

Copy link
Copy Markdown
Collaborator

Report the real model decision on orchestrator runs instead of the unused binding fallback.

Before: [switchboard] decision: … model=claude-sonnet-4-6 (binding)         ← a model no task runs
After:  [switchboard] decision: … model=chosen-per-task (agent-prompts)
        [orchestrator] task models: seed=openai/gpt-5.6-terra install=openai/gpt-5.6-luna init=openai/gpt-5.6-luna identify=openai/gpt-5.6-terra error-tracking=openai/gpt-5.6-luna capture=openai/gpt-5.6-terra dashboard=openai/gpt-5.6-luna review=openai/gpt-5.6-terra report=openai/gpt-5.6-luna

The second line is the end decision — every task type's resolved model, logged once the context-mill agent prompts load (their model_pi/model_sdk fields decide, with CLI/flag overrides and the switchboard fallback applied), and captured as an orchestrator task models event. Runs with a genuinely pinned run-level model are unchanged.


Created with PostHog Code

…nstead of the unused binding model

An orchestrator run whose model traced to the binding never runs that
model — every task picks its own from prompt frontmatter. The event and
log line said model=claude-sonnet-4-6 (binding) anyway, which keeps
reading as a routing bug during release monitoring.

Generated-By: PostHog Code
Task-Id: affdea28-7761-4d50-8335-6b32376b1a82
@github-actions

Copy link
Copy Markdown

🧙 Wizard CI

Run the Wizard CI and test your changes against wizard-workbench example apps by replying with a GitHub comment using one of the following commands:

Test all apps:

  • /wizard-ci all

Test all apps in a directory:

  • /wizard-ci basic-integration
  • /wizard-ci mcp-analytics
  • /wizard-ci revenue
  • /wizard-ci self-driving

Test an individual app:

  • /wizard-ci basic-integration/android
  • /wizard-ci basic-integration/angular
  • /wizard-ci basic-integration/astro
Show more apps
  • /wizard-ci basic-integration/django
  • /wizard-ci basic-integration/fastapi
  • /wizard-ci basic-integration/flask
  • /wizard-ci basic-integration/javascript-node
  • /wizard-ci basic-integration/javascript-web
  • /wizard-ci basic-integration/laravel
  • /wizard-ci basic-integration/next-js
  • /wizard-ci basic-integration/nuxt
  • /wizard-ci basic-integration/python
  • /wizard-ci basic-integration/rails
  • /wizard-ci basic-integration/react-native
  • /wizard-ci basic-integration/react-router
  • /wizard-ci basic-integration/sveltekit
  • /wizard-ci basic-integration/swift
  • /wizard-ci basic-integration/tanstack-router
  • /wizard-ci basic-integration/tanstack-start
  • /wizard-ci basic-integration/vue
  • /wizard-ci mcp-analytics/custom-dispatcher
  • /wizard-ci mcp-analytics/typescript-sdk
  • /wizard-ci revenue/stripe
  • /wizard-ci self-driving/astro
  • /wizard-ci self-driving/fastapi
  • /wizard-ci self-driving/nuxt
  • /wizard-ci self-driving/react-router
  • /wizard-ci self-driving/sveltekit

Results will be posted here when complete.

Generated-By: PostHog Code
Task-Id: affdea28-7761-4d50-8335-6b32376b1a82
…eed time

The switchboard line only says the model is chosen per task; the end
decision now lands right after the agent prompts load — one log line and
one event with every task type's resolved model.

Generated-By: PostHog Code
Task-Id: affdea28-7761-4d50-8335-6b32376b1a82
@gewenyu99
gewenyu99 marked this pull request as ready for review July 29, 2026 14:34
@gewenyu99
gewenyu99 requested a review from a team as a code owner July 29, 2026 14:34
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants