handover-with-tasks
Skill1 starsChanged 7 months agoMITView on GitHub
Write a handover document that includes an open task list for the next agent. Use when switching sessions mid-way through a multi-task workload, not just a single in-progress job. Triggers on phrases like "handover with tasks", "task handover", "hand off the task list", "context cleanup with tasks", "pass on the work queue".
In this repository13 other files it gives its agents.
- claude-and-shellSkill
- claude-gridSkill
- clipboard-handoverSkill
- handoverSkill
- new-claude-atSkill
- new-claude-hereSkill
- new-workspaceSkill
- sideclaudeSkill
- spawn-planning-repoSkill
- start-from-handoverSkill
- start-leftoverSkill
- terminal-hereSkill
- vscode-hereSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
Before writing, gather ALL of the following:
- Distinctive: line 21
Read CLAUDE.md: if present — understand project conventions
- Very specific: line 26
If $ARGUMENTS contains additional context about the task list, incorporate it.
- Very specific: line 30
Write HANDOVER.md at the repository root with the following structure:
- Very specific: line 90
Ask if they want to spawn a new Claude session to pick up from the handover (using session-transfer:new-claude-here or …
≈ “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.
Files
Handover With Task List
A 2-in-1 handover skill: writes a structured HANDOVER.md that captures both the current state AND an actionable task list for the next agent. Use this instead of the regular handover when the session has an open queue of work items rather than a single focused task.
When to use this vs regular handover
- Regular handover (
session-transfer:handover): One task in progress, next agent picks up where you left off - This skill: Multiple tasks, some done, some in progress, some not started. The next agent needs a prioritized work queue, not just resumption instructions
Investigation Process
Before writing, gather ALL of the following:
- Git status — staged/unstaged changes, current branch, recent commits (
git log --oneline -20) - Git diff — what's changed since last commit
- Read CLAUDE.md if present — understand project conventions
- Scan for existing plans/tasks — TODO files, task lists, issues, any planning docs
- Review conversation context — what tasks were discussed, requested, completed, deferred
- Identify the active work area — which files were most recently modified
If $ARGUMENTS contains additional context about the task list, incorporate it.
Output
Write HANDOVER.md at the repository root with the following structure:
# Agent Handover Document
## Metadata
- **Created**: [ISO 8601 timestamp]
- **Creating agent**: [Agent type and model]
- **Repository**: [repo name]
- **Branch**: [current branch]
- **Last commit**: [short hash + message]
- **Handover type**: Task list transition
## Context
[Brief summary of what this session was doing and why the handover is happening. Include any relevant background the next agent needs to understand the overall objective.]
## Task List
### Completed
- [x] [Task description] — [brief note on outcome/location of work]
### In Progress
- [ ] [Task description] — [current state, what's been done, what remains]
- **Files touched**: [list]
- **Current state**: [where exactly this was left]
- **Next action**: [specific next step]
### Not Started
- [ ] [Task description] — [any context or prerequisites]
### Deferred / Out of Scope
- [ ] [Task description] — [why deferred, any notes for future]
## Current Repository State
- **Build status**: [passing/failing/unknown]
- **Uncommitted changes**: [yes/no — summarize if yes]
- **Working tree clean**: [yes/no]
## Key Decisions Made
- [Decision]: [Rationale]
## Failed Approaches (Do Not Repeat)
- [What was tried]: [Why it failed]
## Files of Interest
| File | Why |
|------|-----|
| [path] | [reason this file matters] |
## Resumption Instructions
[Tell the next agent to start by reviewing the Task List above and begin with the highest-priority incomplete item. Include any ordering preferences, dependencies between tasks, or constraints.]
## Context the Next Agent Needs
[Environment variables, running services, non-obvious codebase quirks, etc.]After Writing
- Confirm HANDOVER.md was written successfully
- Give the user a summary: N completed, N in progress, N remaining
- Ask if they want to spawn a new Claude session to pick up from the handover (using
session-transfer:new-claude-hereorsession-transfer:handoverspawn pattern)
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.

