Skip to content

[epic] Business feature implementation status tracker (#1 SHIPPED / #2 VISION / #3 PARTIAL) #335

Description

@LiuShiyuMath
  ____                  _                    ____                                _
 | __ ) _   _ ___(_)_ __   ___  ___ ___    |  _ \ ___  __ _ ___  ___ ___      _| |
 |  _ \| | | / __| | '_ \ / _ \/ __/ __|   | |_) / _ \/ _` / __|/ __/ _ \    / _` |
 | |_) | |_| \__ \ | | | |  __/\__ \__ \   |  _ <  __/ (_| \__ \ (_| (_) |  | (_| |
 |____/ \__,_|___/_|_| |_|\___||___/___/   |_| \_\___|\__,_|___/\___\___/    \__,_|
                                                                            tracker
  #1 SHIPPED · #2 VISION · #3 PARTIAL    canonical: docs/BUSINESS-FEATURES.md

Business Feature Implementation Status — Epic Tracker

Single tracking issue for the three canonical business features defined in
docs/BUSINESS-FEATURES.md. Status
columns mirror docs/verify/BUSINESS-FEATURE-HARNESS-MAP.md.

Per docs/TRIAGE-AND-SPLIT.md: this is an epic tracking issue created with
the epic label from inception (not retroactively bolted on). Child issues
already exist; new sub-issues should follow ≤50-word FIXEDFLOW template +
grill cycle.


Canonical anchor (verbatim)

TeamBrain has three business features: (1) new Claude Code instances no longer make mistakes that previous Claude Code has made; (2) team leaders know in second-level realtime what each teammate's Claude Code instance is doing; (3) video recording and uploading to centralized data storage is easy to use.


Feature-level status

# Feature Status Third-party harness LLM-cannot-fake?
1 no longer make mistakes / previous Claude Code SHIPPED (a) docs/plans/2026-05-11-feature1-init-judge/judge.md openable-and-usable gate; (b) docs/verify/E2E-LEARNING.md Counterfactual Ablation (scipy.stats.ttest_rel) + byte-level Regression Replay (ADR-0010/0012) ✅ both gates
2 second-level realtime / teammate's Claude Code instance VISION plan-only: docs/plans/2026-05-11-feature-2-secondlevel-realtime/plan.md n/a — harness lands with impl
3 video recording / centralized data storage PARTIAL transcript-level capture shipped in auto-capture / team-share; video stream + centralized storage turnkey: no plan yet n/a — turnkey not started

Existing open issues attached to this epic

Feature #1 — auto-capture / matcher / init / dogfood (SHIPPED, polishing)

Feature #2 — second-level realtime dashboard (VISION → plan in flight)

Feature #3 — video recording + centralized storage (PARTIAL)

Cross-cutting (Codex parity / install propagation — supports #1 + #2 + #3)


Sub-issue TODO list (to-be-created child issues)

Below is the suggested split for each feature into FIXEDFLOW-ready child issues
(each ≤50-word body + own grill cycle). Maintainer should grill-review before
running /fixed-flow-driver
per docs/TRIAGE-AND-SPLIT.md.

Feature #1 — close the loop on what's already shipped

Feature #2 — turn VISION into a thin slice

Feature #3 — open a real plan for the video+storage turnkey

Cross-cutting


Coordinator + boundaries

  • Coordinator: maintainer @LiuShiyuMath (until reassigned in this issue's comments)
  • PR-N boundaries: each child issue above ships as exactly one squash-merged PR; this epic does not ship code itself
  • Cadence: per docs/PUBLISHING.mdPublish per 10 PRs; this epic spans ≈3 release branches

Source of truth

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    epicTracking issue split into child issues per docs/TRIAGE-AND-SPLIT.md

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions