--- orphan: true --- # Watch: E2E stack — {{TOPIC}} ({{EFFORT}}) **Date:** {{YYYY-MM-DD}} **Result:** in progress | completed | blocked ## Plan | Field | Value | | --- | --- | | Topic | `{{TOPIC}}` | | Effort | `{{EFFORT}}` (`low` ~25 rows, `med` ~250, `high` 2-hop ~3k) | | Collection | `e2e/{{TOPIC}}` (or `-C` override) | | Years | `{{YEAR_FROM}}`–`{{YEAR_TO}}` (default: previous → current calendar year) | ## Run - Terminals: `` (first pass), `` (resume if any) - cwd: repo root - Mode: start + watch - Started watching: {{LOCAL_START}} - Finished: (when done) ```sh PAPERFUL_E2E=1 uv run python scripts/e2e_stack.py \ --topic "{{TOPIC}}" \ --effort {{EFFORT}} \ {{YEAR_FLAGS}} # resume example: PAPERFUL_E2E=1 uv run python scripts/e2e_stack.py \ --topic "{{TOPIC}}" --effort {{EFFORT}} \ --from-phase --run-id ``` Goal: watch until the harness exits. On failure or hang, fix, prove with a focused test, then continue with `--from-phase` / `--run-id` (same topic and effort). ## Log - Seeded assessment; starting live run. - Run id: `` from harness stdout / `state/e2e//`. - (append phase milestones, incidents, fixes) ## Incidents | Kind | Detail | | --- | --- | | Failure | | | Hang | | | Kill | | ## Fixes 1. (minimal code changes + pytest proof) ## Artifacts - `state/e2e//report.json` / `REPORT.md` - `state/e2e//reachout.csv` - Zotero `-C e2e/{{TOPIC}}`