Compare displayed and user-expected statuses from the authorized live Yonsei attendance page first, identify records explicitly disputed by the user, and report which items have enough reason and evidence to draft a correction; use a screenshot, pasted table, export, or JSON only when the live page is unavailable or intentionally supplied. Use without inferring presence or changing official records.
Scanned 9/1/2026
Install to Claude Code
npx -y skills add mrcha033/yonsei-skills --skill find-yonsei-attendance-discrepancies --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Find Yonsei Attendance Discrepancies?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/mrcha033-find-yonsei-attendance-discrepancies)More formats (shields.io, HTML) on the badges page.
---
name: find-yonsei-attendance-discrepancies
description: Compare displayed and user-expected statuses from the authorized live Yonsei attendance page first, identify records explicitly disputed by the user, and report which items have enough reason and evidence to draft a correction; use a screenshot, pasted table, export, or JSON only when the live page is unavailable or intentionally supplied. Use without inferring presence or changing official records.
---
# Find Yonsei Attendance Discrepancies
Return one discrepancy report from a supplied review snapshot.
## Prepare the input
If the user supplies an attendance screen or table, extract the displayed
records into a private temporary JSON file, then ask the user which entries they
dispute and what evidence they want considered. Do not make the user write JSON
or infer an expected status from location or timetable data.
## Run
```bash
python3 "$SKILL_DIR/scripts/find_discrepancies.py" --input attendance-review.json
```
Each record needs course, date, and `recorded_status`. Supply `expected_status` when known, or set `user_disputed: true`. Set `reviewed: true` for a record the user checked and accepts. A discrepancy is ready for a draft only when it has a different expected status, a reason, and at least one evidence description.
`no_discrepancies_found` is true only when every row is explicitly reviewed. Unreviewed rows remain `unknowns`.
## Boundaries
- Never infer presence from timetable, location, Bluetooth, device, or network data.
- Process only user-supplied attachments, pasted data, or JSON and reject credential, session, check-in, beacon, or location fields.
- Do not accuse a person or system of error; report a user-review candidate.
- Never submit a correction or alter an attendance record.
Is this your skill, or is something wrong with this listing? . Author removals are honored within 72 hours.
No comments yet. Be the first to comment!