agentleFS
slashnote-appslashnote-app/claude-code-plugin/skills/note-wrapup/SKILL.md

note-wrapup

Skill1 starsChanged 9 months agoMITView on GitHub

Generate a session summary with handoff notes for next session

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 10 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If found → save its id — will use mcp__slashnote__update_note in Step 3 instead of creating new

    line 32
  2. Distinctive:

    If not found → will create a new note in Step 3

    line 33
  3. Distinctive:

    Read note-loop note if exists → task progress

    line 44
  4. Distinctive:

    Read loop state file if exists → blocked/remaining tasks

    line 45
  5. Distinctive:

    Read recent decision notes (green) → decisions made this session

    line 46
  6. Distinctive:

    Read recent bug notes (pink) → known issues

    line 47
  7. Very specific:

    If Step 0 found an existing wrapup from current session (< 3 hours) → update it with mcp__slashnote__update_note

    line 106
  8. Distinctive:

    Omit: if no decisions were made

    line 120
  9. Distinctive:

    Omit: if no risks identified — don't invent problems

    line 140
  10. Distinctive:

    Don't repeat info across sections

    line 179
  11. Distinctive:

    If user provides extra context → add as first item in relevant section

    line 181
  12. Distinctive:

    If loop was active → include progress in "Done"

    line 182

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/note-wrapup — Session Summary

Generate an end-of-session summary with handoff notes.

Usage

/note-wrapup                       # Full analysis (git + notes + conversation)
/note-wrapup --notes               # Notes-only mode (no git required)
/note-wrapup <additional context>  # Add extra context to the summary

Behavior

Step 0: Find existing wrapup to update

  1. mcp__slashnote__search_notes with query "Wrapup"
  2. Among results, find one where preview starts with "Wrapup" AND updatedAt is less than 3 hours ago
  3. If found → save its id — will use mcp__slashnote__update_note in Step 3 instead of creating new
  4. If not found → will create a new note in Step 3

Step 1: Collect data

Git data (skip if --notes or not in a git repo):

echo "LOG:" && git log --oneline -10 2>/dev/null && echo "---DIFF---" && git diff --stat 2>/dev/null && echo "---STAGED---" && git diff --staged --stat 2>/dev/null && echo "---BRANCH---" && git rev-parse --abbrev-ref HEAD 2>/dev/null

SlashNote data (always collected):

  1. mcp__slashnote__list_notes — overview of all notes
  2. Read note-loop note if exists → task progress
  3. Read loop state file if exists → blocked/remaining tasks
  4. Read recent decision notes (green) → decisions made this session
  5. Read recent bug notes (pink) → known issues

Conversation context (always available):

  • Tasks discussed and completed
  • Decisions made
  • Issues encountered
  • Files modified

Step 2: Generate summary

Full Mode (git + notes)

# Wrapup <date>

## Done
- <completed work — from git commits + note checkboxes>
- <loop progress if applicable: "/note-loop: N/M tasks">

## Decisions
- <decisions made this session — from /note-decide notes>

## Changed
- <file areas modified — from git diff --stat>
- <uncommitted changes warning if any>

## Open
- <uncompleted tasks, open bugs>
- <uncommitted/unstaged changes>

## Risks
- <potential issues for next session>

## Next
- <exact next steps with file:line references>
- <pending reviews or PRs>

Notes-Only Mode (--notes)

# Wrapup <date>

## Done
- <completed items from note checkboxes>
- <decisions made (from decision notes)>

## Open
- <in-progress items from notes>
- <unchecked note-loop items>

## Next
- <specific next steps based on remaining work>

Step 3: Create or update note

  • Color: green
  • Pinned: false
  • If Step 0 found an existing wrapup from current session (< 3 hours) → update it with mcp__slashnote__update_note
  • Otherwise → create new with mcp__slashnote__create_note

Section Details

"Done" — What was accomplished

  • Group related commits (same pattern as /note-standup)
  • Include completed checkbox items from notes
  • Include loop progress: "/note-loop: 9/12 tasks completed"
  • This is a record — only things actually done

"Decisions" — Choices made this session

  • Source from green decision notes created during this session
  • Format: one-liner per decision (title from note)
  • Omit if no decisions were made
  • Cross-reference with git to see if decisions were implemented

"Changed" — Files and areas modified

  • From git diff --stat (both staged and unstaged)
  • Group by directory/module if many files
  • Flag uncommitted changes: "3 files with uncommitted changes"
  • Omit in notes-only mode

"Open" — Unfinished work

  • Uncommitted/unstaged file changes
  • In-progress checkbox items
  • Open bugs (pink notes from this session)
  • Blocked loop tasks (from state file)

"Risks" — Watch out for

  • Uncommitted changes that could be lost
  • Blocked tasks that need external input
  • Failing tests mentioned in conversation
  • Known bugs that could affect next work
  • Omit if no risks identified — don't invent problems

"Next" — THE handoff section

This is the most important section. It must be specific enough for a fresh session to pick up immediately.

Quality bar — each item should answer: "What file, what function, what's left to do?"

Good:

  • "Continue validateToken() in auth.swift:45 — JWT parsing done, need expiry check"
  • "Run tests after fixing flaky WebSocket test in ws.test.ts"
  • "PR #42 needs one more review — Alice approved"

Bad:

  • "Keep working on auth"
  • "Fix bugs"
  • "Continue development"

Analysis Intelligence

Cross-reference data sources for richer summary:

Cross-ReferenceIntelligence
Git commits ↔ Note checkboxesIf commit matches checkbox → confidently "Done"
git diff ↔ Open tasksModified files show WIP areas
Decision notes ↔ CommitsDecisions that were already implemented
Bug notes ↔ Open itemsKnown issues that block next steps
Loop state ↔ /note-loop noteAccurate progress tracking
Staged vs unstagedWhat's ready to commit vs still WIP

Rules

  • One Bash call max for git data
  • "Done" = actual work (verifiable from git/notes), not plans
  • "Changed" = actual files from git diff --stat
  • "Open" = truly open work, not general TODOs
  • "Risks" = real identified risks, not hypotheticals
  • "Next" = specific, actionable, with file/function references
  • 3-5 items per section max
  • Don't repeat info across sections
  • Date: "Feb 19, 2026"
  • If user provides extra context → add as first item in relevant section
  • If loop was active → include progress in "Done"

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.