Repository navigation
Conversation
…B-227 §5 reconciled) Implementation plan approved 2026-11-03: dashboard, Studio replay and the Python scripts read the bag natively (vendored @mcap/core + fzstd, JS port of rig_nodes.cdr, one shared Python loader); run bags published from the rig PC into arena-experiments with a per-run index fragment; two logging options remain (MCAP default on the rig/ graph, behavior_v2 JSONL on the classic bridge; behavior_v1 and full retired); PR order in three candidates; open questions for Michael listed separately. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Implementation plan for LAB-309 (dashboard, Studio replay and the Python scripts read the MCAP bag; the behavior_v2 JSONL retired as the default), reconciled with LAB-227 §5 (run logs published from the rig PC into
reiserlab/arena-experiments, bag location and index in the repo). Approved by Michael on 2026-11-03 with one direction change: two logging options remain — MCAP by default on therig/graph, the maturebehavior_v2JSONL for the classic FicTrac bridge setup (committed by the Studio as today);behavior_v1andfullare retired.The document answers: (1) which readers read MCAP natively (all browser readers, via a vendored
@mcap/core+fzstdand a JS port ofrig_nodes.cdr; the Python scripts through the converter's record generator behind one shared loader); (2) where bags live (run bag in the repo beside a.meta.jsonfragment, media on the group share, indexformat_version: 2with outcome / run_type / format / media); (3) the rig-PC publisher (rig/nodes/publish_run.py, Contents API, rig-PC PAT now, GitHub App later) and how the Studio's JSONL auto-commit switches off per rig throughhello_ack.publisherwithout touching classic benches; (4) the historical JSONL as a read-only archive; (5) the PR order in three next-candidates with tests. Open questions for Michael are listed separately (Part D).Linear: LAB-309, LAB-227. The plan proposes a scope rewrite of LAB-309's "Done when" (document §B5, last paragraph); not applied.
Tier
docs/skills-only.
Release notes
none
Test plan
🤖 Generated with Claude Code