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.
git clone --depth 1 https://github.com/yongjip/mergetrainWrote 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/plugins/yongjip/mergetrain/plugin)<a href="https://agentmods.dev/plugins/yongjip/mergetrain/plugin"><img src="https://agentmods.dev/badge/plugins/yongjip/mergetrain/plugin/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/plugins/yongjip/mergetrain/plugin"><img src="https://agentmods.dev/badge/plugins/yongjip/mergetrain/plugin.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
mergetrain 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 today.
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.
What it actually says
{
"$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
"name": "mergetrain",
"displayName": "mergetrain",
"version": "3.5.1",
"description": "Queue committed branches from parallel LLM coding agents in Git worktrees. Test the combined tree in a local merge queue, require approval for Git pushes, and recover interrupted pushes.",
"author": {
"name": "Yongjip Kim",
"url": "https://github.com/yongjip"
},
"homepage": "https://yongjip.github.io/mergetrain/",
"repository": "https://github.com/yongjip/mergetrain",
"license": "MIT",
"keywords": [
"git",
"git-worktree",
"worktree-workflow",
"branch-integration",
"merge-queue",
"local-merge-queue",
"git-merge-queue",
"mcp",
"coding-agents",
"llm-agents",
"parallel-agents",
"claude-code",
"codex",
"push-recovery"
]
}
What it installs
The manifest is a name and a version. 2 skills, 1 MCP server travel with it, and installing the plugin installs all of them — 105 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- today Changed 4b49431a4270
- 4d ago Changed · +6 lines 04f2f03b35bd
- 13d ago Changed 30673345fed9
- 16d ago Changed 8b06a8ddd870
- 17d ago Changed eb79b7f1ef8d
- 1mo ago Changed fec34116af3f
- 1mo ago Changed adaf0bd8fc3a
- 1mo ago Changed · +4 lines 27140194af2f
- 1mo ago First seen · 21 lines scan A 85007b2b4974
mergetrain is a plugin published in the GitHub repository yongjip/mergetrain (4 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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.
Other plugins, from other repositories
git-sidebar
A lazygit-style side pane: every worktree of the repository and its local branches as rows you click. A worktree row moves the session there with /cd; a branch row shows its last commits and offers switch (clean tree only) or open worktree. Docked floor to ceiling beside the transcript in the fullscreen layout, above…
git
Git workflow tools: commits, PRs, review inbox, checkout, and work triage.
git-pr
Committing, PR creation and review triage, and dependency upgrades.
cas
Evidence-driven research, guarded Codex delegation, explicit native parallelism, secret protection, and approval gates for rm, commit, and push.
code-quality
OxLint, Biome, and Prettier configuration, linting skills, and pre-commit hooks for consistent code quality.
prek
Set up git pre-commit hooks with prek, the fast pre-commit-compatible runner.