# Repo runtime map

## Existing runtime pieces

- Assistant backend: `llm_normalizer/backend`
- Technical export formatter: `llm_normalizer/frontend/src/utils/conversationExport.ts`
- Async single-case runner: `POST /api/eval/run-async/start`
- Async polling: `GET /api/eval/run-async/:job_id`
- Session store path: `llm_normalizer/data/assistant_sessions/<session_id>.json`
- Session API: `GET /api/assistant/session/:session_id`

## Capture strategy for this repo

Current mutable domain source of truth:
- `docs/orchestration/active_domain_contract.json`

1. Prefer automated capture with:
   - `python scripts/domain_case_loop.py run-case ...`
2. For linked multi-step scenarios, capture with:
   - `python scripts/domain_case_loop.py run-scenario --manifest ...`
3. For full domain pools grouped into several scenarios, capture with:
   - `python scripts/domain_case_loop.py run-pack --manifest ...`
4. For autonomous analyst/coder improvement over a full pack, run:
   - `python scripts/domain_case_loop.py run-pack-loop --manifest ...`
5. If baseline already exists as copied markdown export, import it with:
   - `python scripts/domain_case_loop.py import-export ...`
6. Use `baseline_turn.json` / `rerun_turn.json` as canonical analyst input for case mode.
7. Use `scenario_state.json` plus per-step `turn.json` as canonical analyst input for scenario mode.
8. Use `pack_state.json` plus per-scenario `scenario_state.json` as canonical analyst input for pack mode.
9. Use `loop_state.json` plus per-iteration `analyst_verdict.json` / `coder_result.json` as canonical analyst input for autonomous pack-loop mode.
10. Use `baseline_output.md` / `rerun_output.md` or per-step `output.md` as human-readable paired artifacts.
11. For follow-up-heavy domains, use `scenario_acceptance_matrix.md` as the canonical coverage view for scenario-tree nodes, edges, and paraphrase families.

## Default run assumptions

- backend URL: `http://127.0.0.1:8787`
- eval target: `assistant_stage1`
- single-case async run uses generated case id `AUTO-001`
- artifact root: `artifacts/domain_runs/<case_id>/`
- scenario capture uses `POST /api/assistant/message` and `GET /api/assistant/session/:session_id`
- live runners perform backend preflight via `GET /api/health`
- `run-pack-loop` defaults to `gpt-5.4` for analyst and `gpt-5.4-mini` for coder

## Important constraints

- Reuse current assistant runtime; do not build a parallel execution lane.
- Preserve UTF-8 without BOM for every generated artifact.
- Do not overwrite existing AGENTS rules; extend them.
- Do not treat a root node success as domain acceptance when selected-object or drilldown edges on the primary user path are still broken.