McRitchie Studio

← All Tasks

Sizing

Reviewer-select record by default

Archived

reviewer-select-record-by-default

Created

Jun 25, 05:49

Started

Jun 25, 05:50

Completed

Jun 25, 12:23

DevOps handoff

Type

Feature

Shape

backend

Worktree Slug

reviewer-select-record-by-default

Repositories

mcritchie-studio

Release Train

—

Branch

feat/reviewer-select-record-by-default

Pull Request

https://github.com/amcritchie/mcritchie-studio/pull/190

Local URL

http://localhost:3035

QA URL

—

Production URL

—

devops review-routing

Acceptance Criteria

reviewer-select records review-pair intent by default
a no-record flag preserves advisory-only behavior
recorded intent populates board live crew ticker

Expected Test Plan

[unit] reviewer-select records by default, --no-record suppresses
[integration] default run posts the reviewed intent

Checks Run

[review] Jasper (heavy) APPROVE + Shannon (light) APPROVE — both suites + failure-injection green (15/79 + 38/176); default posts exactly 1 reviewed-intent, --no-record/--dry post 0, best-effort write safe, docs consistent

Agent Context

From conductor retro 2026-06-25. bin/reviewer-select currently only POSTs the review-pair intent when run with --record (the POST is at ~L280); conductors run it read-only to get the assignment, so no intent TaskEvent is created and the board's live deploy-lane crew ticker (data-test=crew-live, stage_agents_helper#in_progress_work) stays empty — operator flagged 'I don't see agent intent working'. Make recording the DEFAULT when a task slug is given: reviewer-select <slug> posts the --to reviewed intent with the chosen heavy/light pair. Add --no-record (or --dry) for the advisory-only/preview case (must NOT post). Keep the auditable seeded-tiebreak output. See feedback_record_intent_on_pr_review. Test: unit for flag parsing/default; integration for the intent POST happening by default and suppressed under --no-record.

Stage Timeline

Who handled each stage, the time it took (measured), and the model / tokens / cost reported (best-effort) — plus who's on it right now. — means the agent didn't report that metric.

Created → Designed

N
Ninetales

Model

—

Duration

—

Tokens

—

Cost

—

Completed Jun 25, 05:49 · 2 days ago

api
Designed → Building

N
Ninetales

Model

claude-opus-4-8

Duration

2 minutes

Tokens

—

Cost

—

Started Jun 25, 05:49

Completed Jun 25, 05:50 · 2 days ago

cli
Building → Submitted

N
Ninetales

Model

claude-opus-4-8

Duration

11 minutes

Tokens

—

Cost

—

Started Jun 25, 05:50

Completed Jun 25, 06:01 · 2 days ago

cli
Submitted → Reviewed

J
Jasper 2

S
Shannon 1

Model

claude-opus-4-8

Duration

about 6 hours

Tokens

26,758,866

Cost

~$25.18

Started Jun 25, 06:01

Completed Jun 25, 12:03 · 2 days ago

cli
Reviewed → Assembled

S
Steffon

Model

—

Duration

6 minutes

Tokens

—

Cost

—

Started Jun 25, 12:03

Completed Jun 25, 12:09 · 2 days ago
Assembled → Shipped

A
Avi

Model

—

Duration

14 minutes

Tokens

—

Cost

—

Started Jun 25, 12:09

Completed Jun 25, 12:23 · 2 days ago
Shipped → Archived

8
81fad33e-98b5-4075-8ed6-d0efe6cb5804

Model

claude-opus-4-8

Duration

about 9 hours

Tokens

—

Cost

—

Started Jun 25, 12:23

Completed Jun 25, 21:16 · 1 day ago

cli