Skip to content
SkillModel-led257 tokens in use

claude-handoff

Hand the current conversation to a fresh named background Claude agent that picks up the work immediately.

328,764 stars on the repository MIT Licensemattpocock/skills · skills/in-progress/claude-handoff
claude plugin install mattpocock-skills@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing claude-handoff costs

22tokens idle
257tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts and code blocks. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Needs attention
  • Frontmatter8/15
  • Structure3/25
  • Examples0/15
  • Output7/15
  • Size15/15
  • Resources0/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community56
  • Maturity32
  • Momentum40

Claude handoff writes a summary of the current conversation, saves it to the operating system’s temporary directory and starts a background agent seeded with that file as its prompt. The agent starts in the current working directory and returns at once, and you manage it with claude agents.

The summary names the skills the next agent should call and points to specs, plans, issues and commits by path or URL.

What You Get

  • A handoff summary file in the temp folder
  • A background agent started with a descriptive name that shows in the job list and terminal title
  • A suggested skills section for the next agent
  • A summary with secrets and personal information redacted

Common Use Cases

  • Continue a task in a fresh context window without losing the thread
  • Spin off a follow-up job such as a bug fix while you keep working
  • Pass work to a new agent session with a focused brief

What Makes It Unique

It launches the next agent itself, so you get a running session and no document to paste. It passes the summary through a file so the shell does not expand backticks or dollar signs in the text. It also avoids duplicating specs, plans, ADRs and diffs by pointing to them.

Under The Hood

A short SKILL.md with no scripts. The agent writes the summary to the temp directory and runs claude —bg —name with the file contents as the prompt, which starts a local background session. It redacts API keys, passwords and personal data before the summary becomes the prompt.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers