jj

A workflow guide for Jujutsu, or jj, a version-control system that records file history and changes without a separate staging area. It covers status, history, differences, commits, rebasing, splitting, squashing, and Git interoperability.

In plain words
What is it for?
Use it to inspect repository state, create and describe changes, rewrite history, resolve conflicts, and work between jj and Git.
Why use it?
It explains how to manage changes safely in repositories that use jj and how its history model differs from Git.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/adampoit/ai/jj
Any agent
npx skills add adampoit/ai --skill jj
Clone the repo
git clone --depth 1 https://github.com/adampoit/ai

Made for: Claude Code, Codex.

Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 943 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00055 $0.00943
Opus 5 $0.00028 $0.00472
Sonnet 5 $0.00011 $0.00189
Haiku 4.5 $0.00006 $0.00094

Measured 2d ago against content hash f4ab7067bed6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

jj scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/jj/SKILL.md · 174 lines

How it starts

The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.

jj

Use jj for version control when a repository has a .jj/ directory or jj status succeeds.

jj status

Core Model

  • There is no staging area; commands operate on the current change or selected revisions.
  • The working copy is a mutable commit/change. Use jj describe to set its message.
  • Change IDs are stable across rewrites; commit IDs change when contents or metadata change.
  • Most history edits are safe because jj records operations in the operation log.

Inspect State

Start with these read-only commands:

jj status
jj log --limit 20
jj diff --git

Inspect a specific revision:

jj show <revset>
jj evolog -r <revset>

Useful revsets:

  • @: current working-copy commit
  • @-: parent of the working-copy commit
  • ::@: ancestors of the working-copy commit
  • @::: descendants of the working-copy commit
  • trunk(): repository trunk/main baseline
  • mine(): changes authored by the current user
  • conflicted(): changes with conflicts
  • mutable(): changes that can be rewritten

Quote revsets in shell commands when they contain special characters:

jj log -r 'trunk()..@'

Make Commits

Describe the current change before creating a new one:

jj describe -m "Explain why this change exists"
jj new

Create a new change from a specific parent:

jj new <revset>

Prefer small, atomic changes. If the working copy mixes concerns, split it before describing or submitting.

jj desc is the common shorthand for jj describe; use either form consistently with nearby docs or commands.

Refine History

Move selected changes from the working copy into its parent:

jj squash

Absorb modifications into matching mutable ancestors:

jj absorb

Split a mixed change:

jj split

Split whole files or filesets non-interactively by passing paths and a message:

jj split -r @ path/to/file1 path/to/file2 -m "Move focused changes into their own commit"
jj split -r @ 'glob:src/**/*.rs' -m "Update Rust sources"

Read the full file on GitHub · 174 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 174 lines · 55 tokens per session scan A f4ab7067bed6

Subscribe to this mod's changes

jj is a skill published in the GitHub repository adampoit/ai (2 stars, last pushed 7d ago), licensed MIT. It adds 55 tokens to every session and 943 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories