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/bendrucker/claudeWrote 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/bendrucker/claude/gitlab)<a href="https://agentmods.dev/plugins/bendrucker/claude/gitlab"><img src="https://agentmods.dev/badge/plugins/bendrucker/claude/gitlab/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/bendrucker/claude/gitlab"><img src="https://agentmods.dev/badge/plugins/bendrucker/claude/gitlab.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
gitlab 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 4d 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.
What it actually says
{
"name": "gitlab",
"description": "GitLab workflow best practices and glab CLI usage",
"author": {
"name": "Ben Drucker",
"email": "[email protected]"
},
"dependencies": ["git", "mod-events"],
"userConfig": {
"reviewers": {
"type": "string",
"title": "Extra reviewers",
"description": "Comma-separated usernames that --bots treats as review bots, beyond the structural bot detection. Empty adds none.",
"default": ""
},
"watch": {
"type": "boolean",
"title": "Watch the merge request",
"description": "Watch the current branch's merge request in interactive sessions and show its state in the prompt footer.",
"default": true
}
}
}
What it installs
The manifest is a name and a version. 3 skills travel with it, and installing the plugin installs all of them — 191 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.
- 4d ago Changed · +14 lines 564862c3b56e
- 6d ago Changed 609a244a33b4
- 8d ago Changed · +1 lines 578375137683
- 1mo ago First seen · 9 lines scan A 9e974023d4e9
gitlab is a plugin published in the GitHub repository bendrucker/claude (18 stars, last pushed today), 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 2026-08-30.
Other plugins, from other repositories
review
PR review, interactive git diff annotation review, and writing style guide.
han-github
Github specific extensions to the Han plugin for agentic planning and coding. Provides skills to post code reviews as PR comments, update a PR description with a consistent format, etc.
fuse-rules
Global rules for all Claude Code sessions: APEX workflow, SOLID/DRY enforcement, project detection, agent delegation, frontend/git standards.
gerrit
An MCP server for interacting with the Gerrit code review system.
odin-git
ODIN workflows for commits, branches, pull requests, review feedback, and issue tracking.
cadence
A plan/execute/verify loop for a single developer in Claude Code, with adversarial review at every gate: plans and diffs get refuted by a fresh Claude subagent or a panel of outside models (OpenAI, Gemini, DeepSeek), not just approved. State lives in files, not the conversation, so you clear aggressively and keep both.