close
Skill8 starsChanged 5 months agoMITView on GitHub
Session close ritual. Writes the closing report and updates the brain context. Use when the user ends a work session — says "close", "wrap up", "end of day", or "done for today".
27 other projects use a version of this.
Adopted by
Most recently taken up by
- dorian-akhs/skillstheir own version, first seen Nov 16, 2026
- timursafiullin/aodtheir own version, first seen Nov 15, 2026
- playingpossumm/loose-endstheir own version, first seen Nov 10, 2026
- vzhukovs/podman-desktop-kittheir own version, first seen Nov 1, 2026
- sophonfinance-wq/finance-automation-portfoliotheir own version, first seen Nov 1, 2026
- PapiScholz/session-closuretheir own version, first seen Oct 17, 2026
Used with
Projects with it also use open, reconcile and 6 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
BlockSkillopen
Open markdown documents in Penpal for collaborative review. Use when you want a human to review a document, plan, or research artifact.
Also in 5 projects that use this
howellsSkillreconcile
Audit a project against current plugin conventions and offer to update structure.
Also in 3 projects that use this
Microsoft AzureSkillvalidate
Run the full CI validation workflow locally before submitting a PR. Use this as a final check to ensure all Azure DevOps CI pipeline steps will pass.
Also in 4 projects that use this
OpenAISkillbootstrap
Project bootstrapping orchestrator for repos that depend on Vercel-linked resources (databases, auth, and managed integrations). Use when setting up or repairing a repository so linking, environment provisioning, env pulls, and first-run db/dev commands happen in the correct safe order.
Also in 3 projects that use this
genomewalkerSkillresume
Resume a thread by loading its ~800-token context capsule
Also in 3 projects that use this
MicrosoftSkillinit
Also in 3 projects that use this
Other versions27 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 27
- Kanevry/session-orchestrator★ 53same name
- wrg32786/aigent-os★ 26same name
- sophonfinance-wq/finance-automation-portfolio★ 16same name
- maxwellsdm1867/wheeler★ 12same name
- snapsynapse/skill-provenance★ 7same name
- BriGadja/iapreneurs-claude-code-kit★ 6same name
- apideck-libraries/api-skills★ 5same name
- Magnus-Gille/claude-skills★ 5same name
- charlie947/skills-for-everyone★ 2same name
- davekindl/skills★ 2same name
17 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
Show its rules and text
We split this skill into the rules it gives an agent
15 rules. 0 are common, 7 are distinctive, and 8 are very specific to this project.
0
Common
Most of the 22 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Read brain/team.md and count the entries under users:.
- Distinctive: line 15
2 or more users (TEAM): run Step 1a to identify the owner first.
- Very specific: line 17
If brain/team.md does not exist, treat it as SOLO.
- Very specific: line 21
Ask: "Who is closing this session?" and match the answer to a handle in brain/team.md.
- Distinctive: line 23
If it matches no handle, stop and ask again — closing under the wrong name overwrites someone else's work, which is exactly what the split …
- Very specific: line 27
Create the <user>/ folder if it does not exist.
- Distinctive: line 31
Create or append to the report path from Step 0/1a:
- Distinctive: line 66
Remove rules below apply to whichever file you are updating.
- Very specific: line 68
Update brain/context.md (or brain/context-<user>.md in per-user mode):
- Very specific: line 86
Never append session narrative or a running log to it — that belongs in notes/daily-summaries/.
- Very specific: line 97
If this session produced something worth remembering across sessions — a correction the user gave you, a preference, a non-obvious project …
- Very specific: line 106
If you edited any shared brain file this session (brain/context.md, brain/contexts/*, agents, hooks, skills), append one dated line per …
- Distinctive: line 121
If a closing report for today already exists (at the same path):
- Distinctive: line 123
Do NOT overwrite earlier sessions.
- Distinctive: line 131
Don't ask for confirmation — just write the report and update context.
“Common” is measured: kept by at least 6 in 10 of the 22 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Session Close
Execute the session close protocol.
Step 0: Solo or team?
Read brain/team.md and count the entries under users:.
- 0 or 1 user (SOLO): the report path is
notes/daily-summaries/closing-DDMMYYYY.md. Skip Step 1a. - 2 or more users (TEAM): run Step 1a to identify the owner first.
If brain/team.md does not exist, treat it as SOLO.
Step 1a: Identify the owner (TEAM only)
Ask: "Who is closing this session?" and match the answer to a handle in
brain/team.md. Skip the question if it is already obvious from the session.
If it matches no handle, stop and ask again — closing under the wrong name
overwrites someone else's work, which is exactly what the split prevents.
The report path becomes notes/daily-summaries/<user>/closing-DDMMYYYY.md.
Create the <user>/ folder if it does not exist.
Step 1: Write the closing report
Create or append to the report path from Step 0/1a:
# Closing [DATE]
## TL;DR
- **Done**: [what completed today]
- **Pending**: [what remains]
- **Next**: [next priority action]
## Details
[Brief summary of main activities — 3-5 sentences max]
## Files Created/Modified
| File | Action |
|------|--------|
| `path/to/file` | Created / Modified / Deleted |
## Key Decisions
| Decision | Why |
|----------|-----|
| [what was decided] | [rationale] |
---
**Session Status**: CompletedStep 2: Update context (MANDATORY)
Team with state_isolation: per-user (check brain/team.md): write your own
state to brain/context-<user>.md (create it if it does not exist yet). Touch the shared brain/context.md only for
shared-header items — this-week priorities you own, or a new cross-person
handoff/blocker. Never edit another operator's context-<handle>.md. The Add /
Remove rules below apply to whichever file you are updating.
Update brain/context.md (or brain/context-<user>.md in per-user mode):
Add
- New priorities or decisions
- Status changes on projects/deals
- New pending tasks
Remove (this is not optional)
| Section | Rule |
| This Week | Check off completed items. Remove items done >3 days ago. |
| Active Areas | Update status if anything changed |
| Quick Reference | Update if key info changed |
Principle: context.md is a SNAPSHOT of current state, not a history log. The history lives in closing reports.
Hard rule — keep context.md lean. Target ~60 lines; hard ceiling ~80 lines / 12KB.
Never append session narrative or a running log to it — that belongs in
notes/daily-summaries/. The same goes for brain/contexts/* files (keep each well
under ~35KB), and for each brain/context-<handle>.md in per-user mode (~35KB).
Two guards enforce this: the size-guard hook blocks an over-ceiling write the
moment it happens, and the brain tripwire (git config core.hooksPath .githooks)
blocks a bloated commit. Don't bypass either — trim instead.
Also update the relevant brain/contexts/*.md file if project statuses changed.
Step 2b: Persist durable learnings (auto memory)
If this session produced something worth remembering across sessions — a
correction the user gave you, a preference, a non-obvious project constraint, a
useful reference — save it to auto memory following the memory skill's house
style (typed topic file + a one-line index entry in MEMORY.md). Skip anything
derivable from the code, git, or CLAUDE.md, and anything that only mattered
today. Most sessions save nothing; that's fine.
Step 2a: Log shared-brain changes (TEAM only)
If you edited any shared brain file this session (brain/context.md,
brain/contexts/*, agents, hooks, skills), append one dated line per change to
brain/changelog.md so teammates see it on their next /start. Create the file
from the template header if it does not exist. Routine per-person work
(your own closing report) does not go here.
Format:
## YYYY-MM-DD — <user>
- `brain/contexts/work.md` — [one-line summary of what changed and why it matters to others]Step 3: Multi-session handling
If a closing report for today already exists (at the same path):
- APPEND as
## Session N: [Topic]. - Do NOT overwrite earlier sessions.
Step 4: Confirm
Tell the user: "Session closed. Report saved in <the report path>."
Don't ask for confirmation — just write the report and update context.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

