# Project Brief

## 1. Product Context

This repository is the MVP foundation for The Mindscape of Bro Tsong.

The current phase is the cognitive model management subsystem MVP, also called `model_library_mvp`.

It supports a future problem-answering / cognitive-processing system that will use selected cognitive models to analyze user inputs.

## 2. First User

The first user is the project owner.

This is an OPC-oriented product workflow.

The system should reduce the burden of managing, testing, and reusing cognitive models as a one-person company.

## 3. Core Need

The project solves a model asset management problem:

- Existing cognitive models are scattered across articles and previous model indexes.
- Model cards need stronger source traceability.
- Some early models require regression testing and stabilization.
- The future question-answering system needs callable model specifications.
- Model and card indexes need script-backed synchronization so the library does not drift into scattered files.

## 4. MVP Focus

The MVP focuses on two models:

- QPI
- Intellectual Archaeology

These two models are used to validate the model extraction protocol.

The current index maintenance rule is documented in `docs/INDEX_MAINTENANCE_PROTOCOL.md`: every asset change synchronizes `models/model_index.json` and `cards/card_index.md`, while handoff and release points require a full rebuild or check.

## 5. Long-Term Direction

Future versions may support:

- 8-10 core models
- 20 extended models
- Model selector
- Multi-lens analysis workflow
- Conflict summarization
- Integrated question-answering system
- Source article integration
- Model regression dashboard

These are not part of v0.1 unless explicitly requested.

## 6. Project Ecosystem

This repository is separate from but related to:

- `knowledge-vault`: upstream source archive and discussion records.
- `ccpe-system`: expert-agent and runtime specification workspace.
- `skills-vault`: canonical source for reusable automation skills.
- `writing-workbench`: deep writing production workspace.
- `video-workbench`: dimensional output workspace.

This project may reference artifacts from those repositories, but its own boundary is the model asset foundation for The Mindscape of Bro Tsong.

## 7. Local CCRA First Review

The project may use Local CCRA review as a file-first first-pass review loop before Web CCRA review.

First-version constraints:

- Local CCRA serves only this project.
- Local CCRA review runs are stored under `local_ccra_reviews/`.
- The reviewer runs through CCPE Agent Runtime in a real child review session.
- Formal local review normally uses the same-child `review_turn` plus `planning_turn` pattern.
- v0.1.2 requires explicit helper/runtime policies for bundle audit profile, gate execution mode, routing diff policy, and lifecycle scan scope.
- Local CCRA output is formal local first review, but it does not replace Owner or Web CCRA judgment.
- The `ccra-local-reviewer` Agent Spec and Runtime protocol belong in `ccpe-system`.

## 8. Round Conductor Trial

The project may use the Round Conductor protocol in `docs/ROUND_AUTOMATION_PROTOCOL.md` for Round 05 and later trial runs.

The Round Conductor is the main Codex session acting as a project-local orchestrator. It may prepare round charters, run Local CCRA passes, extract Chinese findings confirmation items, and implement Owner-accepted repairs.

Constraints:

- Owner / Web CCRA defines the public round goal before the conductor starts.
- Codex must write `findings-confirmation.md` before asking Owner to decide Local CCRA findings.
- Codex may repair only findings recorded in `owner-decision.md` as `accept_for_codex_fix`.
- The protocol does not create a generic Skill or move reviewer governance into this repository.
- Web CCRA remains the formal review lane for key product and lifecycle gates.