close-session
/close-session
Kończy sesję: sprawdza wzorce dnia względem pamięci, proponuje awanse notatek, odświeża MEMORY.md i pisze przekazanie.
- Autor
- awrshift ↗
- Repozytorium
- agent-memory-kit ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 13.11.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 1
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 13.11.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Kończy sesję: sprawdza wzorce dnia względem pamięci, proponuje awanse notatek, odświeża MEMORY.md i pisze przekazanie.
Wyzwalacze z opisu
Oryginalny opis (EN)
End-of-session ritual — audit today's patterns against accumulated memory, propose promotions, refresh MEMORY.md, and write the session handoff. Use when the user says "/memory-kit:close-session", "закрой сессию", "закрываем", "we're done for today", "wrap up".
Jak działa
- Step 1 — Capture what's new
- Step 2 — Audit & promote (the valuable part)
- Step 3 — Refresh the MEMORY.md header
- Step 4 — Write the handoff
- Also check (30 seconds)
- What NOT to do
Podgląd SKILL.md
--- name: close-session description: End-of-session ritual — audit today's patterns against accumulated memory, propose promotions, refresh MEMORY.md, and write the session handoff. Use when the user says "/memory-kit:close-session", "закрой сессию", "закрываем", "we're done for today", "wrap up". allowed-tools: Read, Grep, Glob, Edit(.claude/memory/**), Edit(context/handoffs/**), Edit(knowledge/**) --- # /close-session — the end-of-session ritual You close the session in four steps. Don't just dump — **audit**. Everything you write, you announce briefly; anything promoted to rules/concepts needs the user's verbal "yes" first. **A project may adapt the close.** Its own close list is opened FROM these steps — each item says how a step is done in that project and what the project adds — never run instead of them. A step the list does not name is still done: a close run only from the project's list skips the repeat audit of step 2. (one project's working practice; no count) ## Step 1 — Capture what's new Scan this session for observations worth keeping. Append each to `.claude/memory/MEMORY.md` as a ONE-LINE `[YYYY-MM-DD]`-prefixed entry. Short, scannable, no essays (long rationale belongs in a concept article, step 2). - **File each line under a TOPIC heading** of the patterns section (engineering, process, infra, the domain…) — create the topic if none fits. Never open a heading named after the session or the date (`### s66 (2026-09-25)`, `## Session 65 wrap`): the date lives in each line's prefix, the session narrative lives in the handoff (step 4). The SessionStart hook flags such headings. - **Merge at write time:** before adding a line, grep MEMORY.md for its topic (one or two key words); a line on the same topic is replaced or merged into one (dates joined `[d1→d2]`) — never stacked. (one project's working practice; no count) - **Don't capture what the always-loaded layer already says** (`CLAUDE.md`, rules without `paths:`) or a number whose source of truth lives elsewhere (a price table, a config) — point at the source instead of copying the value. ## Step 2 — Audit & promote (the valuable part) Read the date-tagged entries in MEMORY.md. Look for repetition: **did this pattern appear on 3+ different dates?** Surface 2-4 candidates, be specific: "noticed [date], [date], [date] you said X — codify as a rule or a concept article?" - User says yes → write `knowledge/concepts/<topic>.md` (facts + rationale; frontmatter per `knowledge/index.md` spec) or `.claude/rules/<name>.md` (mechanical always/never constraint; a rule is promoted on a family's 3rd counted occurrence — `reference/review-loop.md` §2 — or on a user's correction). Update `knowledge/index.md` (one line per concept). - Promoted or long-absorbed entries get PRUNED from MEMORY.md — that's how it stays under its caps (180 lines / 32 KB / 3000 chars per line). - Writing a fact into a second place? `reference/doc-governance.md` — one SSOT per fact, and when a fact changes you grep its restatements in the same pass. A product or architecture decision goes to the ledger instead — `projects/<name>/decisions-log.md`, format in `reference/decisions-log.md`. A plan, a spec, a research finding or a QA record is NOT memory: it belongs in that project's folder, per the map in its `README.md`. ## Step 3 — Refresh the MEMORY.md header The header (everything above the first `---`) is «current state of work», at most 3 lines, **REPLACED** every close — never a chronicle of past sessions. Per-session detail lives in the handoff, not stacked in the header. A project may make the header one pointer line to the newest handoff — then it is replaced only when it no longer points right; where the project's session-close document sets the form, it wins. A project that closes many sessions a day may run Step 2's 3-date audit once a day (the first close after a set hour) instead of at every close; its session-close document says so. ## Step 4 — Write the handoff When the project has its own `context/handoffs/HANDOFF-TEMPLATE.md`, copy it → `context/handoffs/<topic>-<YYYY-MM-DD>.md` and fill that template's sections, whatever they are named. Only when there is none, copy the plugin's `templates/HANDOFF-TEMPLATE.md` and fill its five sections (what was done · where things stand · memory updates · next steps · opening note). The SessionStart hook injects the newest handoff next time — this file IS the "remember where we left off" mechanism. With parallel windows, the handoff carries no windows table — one line pointing at the round file that holds each window's state (`reference/window-cycle.md` §7); a hand copy of that state at the close is how a written-but-unsent brief once became «working». ## Also check (30 seconds) - **Experiment hygiene:** any `experiments/*-YYYYMMDD/` older than 30 days → ask: still active or close (distill → delete)? - **Backlogs:** `projects/*/BACKLOG.md` task statuses reflect today's reality — a `DONE` nobody verified is still `IN PROGRESS`. - **Project map:** anything you wrote today into a project folder is a row in that project's `README.md` map, and the map's `Last verified` is today's date. - **Specs vs code:** any spec still `building` after a merge, or any `assumed` spec → run `/memory-kit:code-sync` (it reconciles statuses from repo evidence; this ritual does not). - **Caps:** if MEMORY.md is near any cap, do a deeper prune now, not "next time". - **Begun while there is room to do it whole:** start the close early enough that all four steps — capture, audit, header, handoff — are finished before an automatic compaction would cut into them. This is not a count of the session's context — a session does not count or report its context. A project's own rule on session length wins over this line. ## What NOT to do - Don't promote without the user's yes. Repetition makes a CANDIDATE, not a rule. - Don't stack "Prior session" paragraphs in the MEMORY header — replace it. - Don't group today's lines under a session heading — see Step 1. - Don't write a handoff longer than ~1 screen (about 4,000 characters). It's a note, not a transcript. Session start injects 6,000 characters of it; the section whose heading starts with «Opening» always arrives whole, the rest is cut — so the next session's first move belongs in that section, not in a next-steps list.
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · awrshift/agent-memory-kit | 90 | MIT |
| Soda-null/multi-agent-repo-protocol | 100 | MIT |
| shakacode/agent-workflows | 100 | MIT |
| cristobaltormo/cortex-md | 90 | MIT |
| awrshift/claude-memory-kit | 90 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 874b539110f5 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100wrap-up
Prowadzi koniec sesji: zapisuje wykonaną pracę, zadania otwarte i wnioski do pliku przekazania, a na starcie wczytuje go z powrotem. - 86handoff
Zamyka sesję roli: zapisuje otwarte pytania i decyzje w pamięci, zgłasza potrzebne zadania i zwalnia dzierżawę dyspozytora. - 100close
Zapisuje wyniki sesji w trwałej pamięci notatek i przygotowuje punkt wznowienia na następną sesję. - 90wrap-up
Uruchamia listę zamknięcia sesji: wysyłkę zmian, zapis pamięci i wnioski do poprawy pracy w kolejnych sesjach. - 90wrap
Domyka sesję: wykrywa dotknięte projekty, zapisuje zmiany w WORKBENCH.md, skillach i pamięci trwałej oraz proponuje edycje CLAUDE.md, pytając przed zmianami nieodwracalnymi. - 100session-close
Kończy sesję Claude tak, by następna mogła ruszyć bez starej rozmowy, pisząc notatkę przekazania i prompt.