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.
npx skills add jamestexas/agents --skill jj-stackgit clone --depth 1 https://github.com/jamestexas/agentsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jamestexas/agents/jj-stack)<a href="https://agentmods.dev/skills/jamestexas/agents/jj-stack"><img src="https://agentmods.dev/badge/skills/jamestexas/agents/jj-stack/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jamestexas/agents/jj-stack"><img src="https://agentmods.dev/badge/skills/jamestexas/agents/jj-stack.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00124 | $0.03966 |
| Opus 5.5 | $0.00050 | $0.01586 |
| Sonnet 5.5 | $0.00025 | $0.00793 |
| Haiku 4.5 | $0.00012 | $0.00397 |
Grade C, and why
jj-stack scanned grade C with 1 finding 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 1mo 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
cd <main-repo> && jj git init --colocate # non-destructive; git untouched; rm -rf .jj reverts How it starts
The opening of the file, as written. The whole thing — 282 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/jj-stack — split one branch into signed, stacked PRs on a feature branch
The workflow, and why it exists: agentic work wants one branch you (and the
agent) reason about linearly; reviewers want small PRs; the org wants one
clean merge. jj resolves the tension — you develop on one branch, jj split
it into tiny commits, push each as a small PR that targets a feature branch,
and the feature branch is a single squash/merge to main. A reviewer can review
each small PR into feature; the feature merge is the one that hits main.
jj gives the primitives (split, duplicate, bookmark, push, signing backends).
This skill supplies the two things it does not: keyless-signing wiring and
the feature-branch orchestration. Nothing here shells out from a script — it's a
procedure of jj/git/gh commands; the only file written is the jj/git config.
Phase 0 — prerequisites (one-time per machine/repo)
-
jj installed.
command -v jj || brew install jj(formula isjj). -
Colocate the repo. jj cannot
git init --colocateinside a git worktree — do it in the MAIN checkout:cd <main-repo> && jj git init --colocate # non-destructive; git untouched; rm -rf .jj revertsRun it in a linked worktree and jj refuses with a hint that names the two ways out:
Error: Cannot create a colocated jj repo inside a Git worktree. Hint: Run `jj git init` in the main Git repository instead, or use `jj workspace add` …If the target branch is checked out in a linked worktree, that's fine — you duplicate it here (Phase 2) and leave the worktree's branch as the backup.
-
Author identity (empty identity can't push):
jj config set --user user.name "<Name>" jj config set --user user.email "<email>" -
Snapshot noise — and why it is dangerous for agents. jj refuses to snapshot files over ~1 MiB. The refusal is a
Warning:, not an error: the command still exits 0, and the oversized file is silently left untracked and out of the commit you just made. An agent that trusts exit codes will report a clean commit that is missing a file.Warning: Refused to snapshot some files: big.bin: 8.0KiB (8192 bytes); the maximum size allowed is 1.0KiB (1024 bytes)So: grep the output for
Refused to snapshotafter any command that snapshots (status,commit,split,diff— any of them). Then fix the cause by ignoring the cruft (*.test,*.log, build-binary paths in.git/info/exclude) and/or raising the guard for this repo:jj config set --repo snapshot.max-new-file-size 60MiB
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.
- 1mo ago Changed · +113 lines 4c5887524559
- 1mo ago First seen · 169 lines · 124 tokens per session scan C c40c9689e710
jj-stack is a skill published in the GitHub repository jamestexas/agents (2 stars, last pushed 23d ago), licensed Apache-2.0. It adds 124 tokens to every session and 3,966 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-11.
Other skills, from other repositories
commit-messages
Write git commit messages and pull request titles in the WARP format used by the swarms repo.
setup-pre-commit
A setup guide for Git pre-commit checks, which run automatically before a commit is created. It configures Husky, lint-staged, Prettier, type checking, and tests in a repository.
re0-git
Rewrite a finished commit's message into a clean, handoff-ready form in your own log style, so git log alone tells the story. User-invoked: run it after a commit.
commit-and-push
Create a well-formatted git commit and push to remote repository.
customize-contribute-back
Contribute a user's AIWG customization back upstream as a PR — reviews for general applicability, creates branch, opens PR.
effect-ledger
Record an effect, sign an effect, reconcile an effect or check "did this PR merge" through the signed AIWG effect ledger, using one aiwg effect command with scriptable exit codes.