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 agentmods add skills/coboljunkie/gitlab-mcp/gitlabnpx skills add CobolJunkie/gitlab-mcp --skill gitlabgit clone --depth 1 https://github.com/CobolJunkie/gitlab-mcpWhat 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 | $0.00081 | $0.00598 |
| Opus 5 | $0.00041 | $0.00299 |
| Sonnet 5 | $0.00016 | $0.00120 |
| Haiku 4.5 | $0.00008 | $0.00060 |
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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitLab
Use the bundled GitLab MCP tools as the primary interface to GitLab. Use local
git for an existing checkout and use glab only as a narrow fallback when it
is installed and the MCP surface does not cover the operation.
Read references/tool-routing.md before selecting tools.
Resolve the project
- Prefer an explicit GitLab project URL, project ID, or
namespace/projectpath from the user. - For a local checkout, inspect
git remote -vand normalize SSH or HTTPS remotes to a GitLab project path. - If the project is still ambiguous, use
search_projects. - Respect
GITLAB_URLfor self-managed or dedicated GitLab instances. Do not assume GitLab.com.
Route specialized workflows
- For unresolved merge-request feedback, use the sibling
gl-address-commentsskill. - For failed merge-request pipelines, use the sibling
gl-fix-ciskill. - For committing, pushing, and opening a draft merge request, use the sibling
gl-publishskill.
General workflow
- Use
get_gitlab_capabilitieswhen availability, version, licensing, permissions, token scope, or instance configuration may explain an outcome. - Read the current GitLab state before proposing or making a change.
- Use the smallest tool that provides the required data.
- Summarize relevant findings, including project, issue, merge request, branch, pipeline, or job identifiers.
- If the user requested a change, make only the scoped change and then read back the resulting state when practical.
- Report links or stable identifiers so the user can verify the result.
Write safety
- Treat tools marked read-only as safe inspection.
- Before a write, confirm the target project and object from the conversation and current state.
- Do not delete branches, files, comments, or other objects unless deletion is explicitly requested.
- Do not merge a merge request, retry or cancel CI, approve a merge request, or resolve review discussions unless the user asked for that outcome.
- Never expose access tokens, job tokens, private variables, or unredacted secrets from logs.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 61 lines · 81 tokens per session scan A 75766b339137
gitlab is a skill published in the GitHub repository CobolJunkie/gitlab-mcp (0 stars, last pushed 13d ago), licensed MIT. It adds 81 tokens to every session and 598 once invoked, about $0.0004 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.
Other skills, from other repositories
gitlab-issues
Issue review and tracking on GitLab via the gitlab-api MCP server — review a project's issues, review issues across a whole group/organization, and read, create, update, or close an issue with the domain-typed tool. Use when the agent must triage a project's issues, see every issue affecting a group, open a new issue…
asana-sweep
Scheduled sweep of incomplete Asana tasks assigned to you — triage into auto/ask/skip, dedupe against existing work, then call asana-to-harness to open GitLab issues that trigger an AI coding harness. Built for headless scheduled runs (claude -p); also runs interactively, with --dry-run support.
asana-to-harness
Turn one Asana task into an implementation plan for an AI coding harness, open a GitLab issue with the plan as its description, and leave a trigger comment (e.g. @claude). Used standalone ("turn this Asana task into a harness issue") or invoked by the asana-sweep skill.
gitlab-group
GitLab group operations via API. ALWAYS use this skill when user wants to: (1) list/view groups, (2) create/update/delete groups, (3) manage group members, (4) list subgroups or group projects, (5) share projects with groups.
gitlab-mr
GitLab merge request operations. ALWAYS use this skill when user wants to: (1) list merge requests, (2) view MR details, (3) create new MRs, (4) approve/merge MRs, (5) checkout MR branches, (6) add notes/comments, (7) rebase MRs.
gitlab-wiki
GitLab wiki operations via API. ALWAYS use this skill when user wants to: (1) list wiki pages, (2) read wiki content, (3) create/update/delete wiki pages, (4) upload wiki attachments.