agentsclimarketplace

Gitlab

Skill yuelenghan/orbit/skills/gitlab

An open-source arsenal of reusable skills for AI agents — repo understanding, code review, delivery workflows, security pentests, and plug-and-play integrations with Jira, GitHub, GitLab, Jenkins & more. Drop into Claude Code, Codex, or any skill-aware host.

Install
npx -y skills add yuelenghan/orbit --skill gitlab

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when the user needs GitLab repository, branch, merge request, comment, review, or commit operations through the packaged launcher and local CLI runtime.

SKILL.md

2.9 KB, 641 tokens by cl100k_base, as published. Nobody here has run it

GitLab Skill

Use this skill to operate GitLab repositories, branches, merge requests, reviews, and commits through the packaged launcher and local CLI runtime.

Within Orbit's shared provider framework, this skill is one implementation of the repo capability.

Runtime and execution

Always run the packaged launcher through ${GITLAB_SKILL_DIR} so it works from the user's current working directory:

  • macOS/Linux: ${GITLAB_SKILL_DIR}/bin/gitlab
  • Windows: ${GITLAB_SKILL_DIR}\bin\gitlab.cmd

${GITLAB_SKILL_DIR} is a documentation placeholder, not a guaranteed shell environment variable. Resolve BIN to a real launcher path once before the first CLI call. Do not call node app/dist/cli.js directly.

Scope and boundaries

  • GitLab API through the packaged CLI only.
  • CLI-mediated operations only.
  • Do not call GitLab REST endpoints directly from skill content.
  • Treat merge requests as the repo.pr.* capability surface.

Capability Domain

Domain:

  • repo

Provides:

  • repo.config.get
  • repo.config.set
  • repo.repo.list
  • repo.repo.search
  • repo.repo.get
  • repo.repo.content
  • repo.repo.default-branch.get
  • repo.branch.list
  • repo.branch.create
  • repo.pr.list
  • repo.pr.get
  • repo.pr.changes
  • repo.pr.diff
  • repo.pr.activities
  • repo.pr.create
  • repo.pr.update
  • repo.pr.comment
  • repo.pr.review-comment
  • repo.pr.approve
  • repo.commit.list
  • repo.commit.get
  • repo.commit.changes
  • repo.commit.pull-requests
  • provider-specific workflow helpers

Config check:

  • gitlab config get

Config check

Run gitlab config get when configuration state is unknown or before the first provider-scoped task in a new session.

Quick reference

TaskCommand
Version diagnosticsBIN version
Runtime diagnosticsBIN doctor
Store configBIN config set --base-url <URL> --token <TOKEN>
Read config metadataBIN config get
List namespace reposBIN repo list --namespace <group-or-user>
Get repositoryBIN repo get --namespace <group-or-user> --repo <name>
Read file contentBIN repo content --namespace <group-or-user> --repo <name> --path <path> [--at <ref>]
List merge requestsBIN pr list --namespace <group-or-user> --repo <name>
Get merge requestBIN pr get --namespace <group-or-user> --repo <name> --number <iid>

BIN means the resolved launcher path in the current session.

Supporting files

  • Read references/commands.md for the packaged command surface.
  • Read references/usage.md for execution expectations and shared framework positioning.

What ships with it: 9 files

5.7 KB alongside SKILL.md, 2 of them executable

assets/

bin/

references/

Gives 0 of the 12 instructions most pr commit review skills give in 641 tokens

Counted across 888 of the 1,342 authors here whose files we hold, read 2026-08-07

  • Use conventional commits formatin 127 of 888, across 115 files
  • Keep subject line under 72 charactersin 62 of 888, across 48 files
  • Delete branches after mergein 51 of 888, across 38 files
  • Use imperative mood in subject linein 51 of 888, across 42 files
  • Use imperative mood in commit messagesin 44 of 888
  • Verify directory is ignored before creating worktreein 43 of 888, across 12 files
  • Generate a conventional commit messagein 43 of 888
  • Add unignored worktree directories to gitignorein 42 of 888, across 10 files
  • Make atomic commitsin 39 of 888, across 27 files
  • Run tests before committingin 36 of 888, across 25 files
  • Verify clean test baselinein 35 of 888, across 9 files
  • Split unrelated changes into separate commitsin 35 of 888, across 30 files

Said here and by no other author read

  • resolve the launcher path before the first CLI call
  • run the packaged launcher for all operations
  • run config check when configuration state is unknown
  • run config check before the first provider-scoped task
  • treat merge requests as the repo.pr capability surface
  • read references/commands.md for the command surface

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 327,132. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.