list-learnus-courses
Read the authorized user's live LearnUs dashboard first and return the visible course index as a deduplicated structured list; use a user-supplied screenshot,…
Detect newly listed, changed, and no-longer-visible notices across the current bounded official Yonsei University and Sinchon IT notice windows using a user-selected JSON state file. Use for recurring Yonsei notice checks, monitoring a topic, comparing with a prior run, or
$ npx -y skills add mrcha033/yonsei-skills --skill watch-yonsei-notices --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/watch-yonsei-noticesContext preview
The summary Claude sees to decide when to auto-load this skill.
Detect newly listed, changed, and no-longer-visible notices across the current bounded official Yonsei University and Sinchon IT notice windows using a user-selected JSON state file. Use for recurring Yonsei notice checks, monitoring a topic, comparing with a prior run, or
name: watch-yonsei-notices description: Detect newly listed, changed, and no-longer-visible notices across the current bounded official Yonsei University and Sinchon IT notice windows using a user-selected JSON state file. Use for recurring Yonsei notice checks, monitoring a topic, comparing with a prior run, or building a local read-only notice watch without credentials.
Require the user to choose or approve an explicit state path. Then establish or compare a baseline:
python3 "$SKILL_DIR/scripts/watch_yonsei_notices.py" \ --source all \ --contains "메일" \ --state "/explicit/user-approved/path/yonsei-mail-notices.json"
The first run sets `initialized: true`, writes the baseline, and intentionally returns no `added` alerts. Later runs report `added`, `updated`, and `missing_from_current_window`. Use `--dry-run` to compare without replacing the state.
Read `partial`, `errors`, `source_health_issues`, and `state_write_blocked_reason` before reporting changes. A partial fetch never replaces a healthy baseline. A source that suddenly parses zero items after a healthy non-empty baseline is reported as `suspicious_empty_source`; its baseline is retained because the official page may have changed shape. A genuinely empty first run is still allowed to establish a baseline. Never describe `missing_from_current_window` as deletion: an item can merely have rolled out of a source's bounded recent window.
Keep separate state files for different source, topic, or date filters. The command rejects a changed query for an existing state path; use another path, or pass `--reset` only when the user explicitly asks to replace that baseline.
Do not invent a default state location or persist anything unless `--state` is explicitly supplied. This workflow is public, read-only, and requires neither credentials nor VPN.
연세대학교 생활에서 자주 반복되는 확인·정리 작업을 자연어로 처리하는 비공식 오픈소스 플러그인 모음입니다. 연세대학교 공식 서비스가 아닙니다. 학교 계정 권한을 늘리거나 수강신청·출석·예약 제한을 우회하지 않습니다.
Read the authorized user's live LearnUs dashboard first and return the visible course index as a deduplicated structured list; use a user-supplied screenshot,…
Read an authorized live LearnUs course first and extract deadlines visibly associated with assignment activities; use a user-supplied screenshot, pasted page,…
Read an authorized live LearnUs course first and extract its files, folders, resources, and visible media entries as a structured material list; use a…
Connect to Yonsei LearnUs in the student's persistent browser profile, reuse the official portal login, and recover from session expiry without receiving…
Read the student's authorized Underwood credit-progress table and calculate advisory Yonsei graduation progress from official entry-year, college, and major…
Normalize enrollment and academic-status fields from a user-provided Yonsei Academic Information System screenshot, pasted table, export, or JSON, identify…