메뉴
Safe workflow for changing fuzzy-match threshold or OCR confidence cutoff values. Use whenever editing threshold/score_cutoff in core/item_actions.py or ocr/inventory_vision.py. Corpus replay required before shipping - see T001.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when user wants to Run the Arc Raiders AutoScrapper validation stack after edits. Use for linting, typing, tests, workflow checks, and broader repo validation.
Use when user wants to Update Metaforge snapshots and bundled default rules
Use when user wants to Validate OCR changes against failure corpus before shipping
Add a new OCR regression fixture from an ocr_debug image. Use when a scan misidentified an item and you want to lock in the correct result as a test case.
Use when user wants to Run full validation, push branch, and open a PR with appropriate context notes
Workflow for safely adding, removing, or renaming persisted config fields in config.py. Use whenever editing dataclass fields in src/autoscrapper/config.py. Covers incrementing CONFIG_VERSION, writing a migration function, and validating the round-trip.
| name | threshold-change |
| description | Safe workflow for changing fuzzy-match threshold or OCR confidence cutoff values. Use whenever editing threshold/score_cutoff in core/item_actions.py or ocr/inventory_vision.py. Corpus replay required before shipping - see T001. |
| disable-model-invocation | true |
Changing the fuzzy-match threshold or OCR confidence cutoff affects every item match decision. A wrong value causes false positives (wrong action taken) or missed matches (items skipped). Corpus replay is required before shipping.
This is TODO T001 in the codebase. The hook in settings.json fires a warning whenever you edit a threshold constant.
grep -rn 'FUZZY_THRESHOLD\|score_cutoff\|threshold' src/autoscrapper/core/item_actions.py
grep -rn 'threshold\|THRESHOLD' src/autoscrapper/ocr/inventory_vision.py
Both files share the same constant - change it in one place only.
grep -n 'FUZZY_THRESHOLD|score_cutoff' src/autoscrapper/core/item_actions.py
Record: before = <current_value>
ls ocr_debug/
If empty, run a dry-run scan to populate it:
uv run autoscrapper scan --dry-run
The corpus must not be cleared between now and the replay. See the corpus-replay skill.
uv run pytest tests/autoscrapper/core/test_item_actions.py tests/autoscrapper/ocr/ -v 2>&1 | tee /tmp/threshold_before.txt
Record pass/fail counts from the summary line.
Edit the threshold constant in src/autoscrapper/core/item_actions.py.
uv run pytest tests/autoscrapper/core/test_item_actions.py tests/autoscrapper/ocr/ -v 2>&1 | tee /tmp/threshold_after.txt
diff /tmp/threshold_before.txt /tmp/threshold_after.txt
Apply the corpus-replay skill pass/fail criteria:
uv run ruff check src/ tests/ uv run pytest
threshold-corpus-replay skill - detailed replay procedureocr-reviewer agent - review the surrounding code changesrc/autoscrapper/core/item_actions.py