Session handoff

Skill christopherlouet/claude-base/.claude/skills/session-handoff

Opinionated Claude Code foundation — Explore → TDD → Audit workflow, auto-detected stack presets (nextjs, fastapi, astro, ...), curl | bash install. MIT.

Install

npx -y skills add christopherlouet/claude-base --skill session-handoff

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Context transfer between AI sessions. Trigger when the user wants to save the context, resume a task, or hand off the work to another session.

SKILL.md

3.1 KB, 720 tokens by cl100k_base, as published. Nobody here has run it

Session Handoff

Objective

Produce a git-committable handoff document (.claude/handoff.md) that transfers a work session's state to a different reader: another developer, another agent, another LLM, or a future session on another machine.

Native features first

Claude Code natively covers resuming your own session — do not rebuild that here:

Need	Native answer
Resume my own session	`claude --resume <id>` (full context)
Recap after a break or `/compact`	`/recap`
Persistent preferences/decisions across sessions	auto memory (`~/.claude/memory/`)

This skill's delta is the cross-boundary case: native memory and --resume are personal and machine-local — they do not transfer to a teammate, a CI agent, another tool, or another workstation. A committed handoff.md does.

When to use

Handoff between developers/agents/tools (the reader is NOT this session's owner)
Work continuing on another machine or in an environment without your ~/.claude
Documentation of work in progress that must live IN the repo

Handoff format

Context file: `.claude/handoff.md`

# Session Handoff

**Date:** [YYYY-MM-DD HH:MM]
**Session:** [ID or description]
**Author:** [Human or agent]

## Project context

**Project:** [Name]
**Branch:** [Branch name]
**Current commit:** [Hash]

## Work state

### Done
- [x] [Task 1] - [Detail]

### In progress
- [ ] [Task 3] - [Current state, where it stands]
  - Modified files: [list]
  - Next step: [description]
  - Possible blocker: [description]

### To do
- [ ] [Task 4] - [Description]

## Decisions made

| Decision | Reason | Rejected alternative |
|----------|--------|---------------------|
| [Choice 1] | [Why] | [Other option] |

## Key files

| File | Role | State |
|---------|------|------|
| `src/xxx.ts` | [Description] | Modified / Created / To modify |

## Patterns and conventions discovered

- [Pattern 1 from the codebase]

## Problems encountered

| Problem | Solution/Workaround | Resolved? |
|----------|---------------------|----------|
| [Problem 1] | [Solution] | Yes/No |

## Notes for the next session

[Specific instructions, pitfalls to avoid, points of attention]

## Useful commands

```bash
# To resume
git checkout [branch]
npm test  # Check that everything passes
# Next step: [description]


## Best practices

- Write the handoff DURING the work, not after
- Be specific about files and lines of code
- Document the decisions AND the reasons
- Mention the pitfalls and workarounds discovered
- NEVER assume the reader has this session's context, memory, or machine

## Rules

- ALWAYS create a handoff before handing work across a person/agent/machine boundary
- ALWAYS include the modified files and their state
- ALWAYS document the architectural decisions
- For resuming YOUR OWN session, prefer the native features above — no handoff file needed

Gives 0 of the 12 instructions most agent orchestration skills give in 720 tokens

Counted across 742 of the 995 authors here whose files we hold, read 2026-08-06

run the full test suite after integrating changesin 53 of 742, across 20 files
reference existing artifacts by path or URLin 52 of 742, across 22 files
dispatch one agent per independent problem domainin 50 of 742, across 17 files
verify fixes do not conflictin 45 of 742, across 13 files
include a suggested skills section in the documentin 45 of 742, across 15 files
redact sensitive informationin 41 of 742, across 11 files
save to the temporary directory of the operating systemin 39 of 742, across 9 files
tailor the document to user-provided focus argumentsin 39 of 742, across 9 files
spot check agent changes for systematic errorsin 34 of 742, across 7 files
write a handoff document summarising the current conversationin 31 of 742, across 6 files
assign each agent a specific scopein 23 of 742, across 8 files
provide specific scope and clear goalin 23 of 742, across 5 files

Said here and by no other author read

create handoff before crossing agent or machine boundaries
include modified files and their state
document decisions and their reasons
document discovered pitfalls and workarounds
use native features to resume own session

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.