Habits for long agent sessions: status briefings with proof, follow-up work run in chosen waves with standing rules kept, subagent fan-out from one brief, and one living document per workstream.
Plugin agent-session-craft from marketplace modootoday in modootoday/marketplace. Commands read from the repo's manifest — 23 days ago.
followup-wave-loopRun the remaining work of a long session as waves - list what is left as a numbered table with a recommended order and a mark on items that will be forgotten if deferred, run only the numbers the user picks as one wave, commit in logical units, stop at every approval gate, name what must be deployed
living-doc-syncKeep one living document per workstream (a shared doc, wiki page or published page) current as the work moves - update its checklist, measured results and open decisions in place, restructure sections instead of appending new text below stale text, date every measurement, and return the document's U
repeated-failure-debug-escalationStop a coding session that keeps retrying the same failing approach - log every attempt with its approach and error, count attempts per approach, stop at three, run a separate diagnostic pass that reproduces the failure and lists hypotheses before any new fix, and hand only verified findings back in
session-status-briefingAnswer "where are we" in a long agent session as a short status in four parts - done (with commit hashes and what ran, how much), in progress (subagents, builds), blocked, and the next decision the user owns - in the user's own language, with a check for work beyond the goal; a resumed or compacted
subagent-fanout-briefSplit a large uniform job (dozens of files, packages or documents) across parallel subagents through one shared BRIEF file - the spec written once, chunks of 16 to 25 items with disjoint file ownership, a hard-rules block (no git, no builds or tests until one final check, touch only your own files),
Your own Naver blog, Place, Smart Store, shopping, search ad and news comment data read through the datalab.tools browser extension, plus card news, short-form video, tutorial posts and social copy built in its editors, with no invented numbers and no change made without asking.
31 skills
Numbers that mean what they say: ClickHouse SQL that reads the key and deduplicates correctly, Naver trend data read as ratios, metrics defined once, experiments designed before launch, retention read by cohort, and open survey responses grouped with every response id kept.
16 skills
Reading passages that hit a measured grade level, language drills, and teacher-side checks for coverage, answer keys, lesson timing, study calendars and rule-valid puzzles, with the numbers and logs shown.
14 skills
Growth and community work: store listings written for the people searching that store, Discord servers run with structure and consistent moderation, newsletters readers open and finish, a month of social posts a small team can ship, launch emails that lead with what changed, and sites submitted to Google, Naver and IndexNow engines from a pipeline.
13 skills
Early product-planning steps done carefully: spar on an idea, frame the problem behind a request, synthesize feedback without overstating it, rank a backlog while saying where the ranking is fragile, write a Korean PRD with testable acceptance criteria, and put a decision on one page with the recommendation first.
13 skills
Korean content that reads well and stays true: product copy fixed to the Toss writing principles, Naver blog posts planned for search, persona tone measured with markers, and ghostwriting that never invents the author's facts.
11 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.