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 xonovex/platform --skill gitlab-guidegit clone --depth 1 https://github.com/xonovex/platformWrote 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/xonovex/platform/gitlab-guide)<a href="https://agentmods.dev/skills/xonovex/platform/gitlab-guide"><img src="https://agentmods.dev/badge/skills/xonovex/platform/gitlab-guide/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/xonovex/platform/gitlab-guide"><img src="https://agentmods.dev/badge/skills/xonovex/platform/gitlab-guide.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 | $0.00109 | $0.03173 |
| Opus 5 | $0.00055 | $0.01587 |
| Sonnet 5 | $0.00022 | $0.00635 |
| Haiku 4.5 | $0.00011 | $0.00317 |
Grade A, and why
gitlab-guide 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitLab workflow delivery and enforcement: quick reference
How to realize tickets, kanban, merge-request delivery, durable context projections, and provider-native external enforcement on GitLab. The operations map finished artifacts and workflow intent onto GitLab-native resources.
The one thing to internalize: a project issue/work item, an issue-board list, a top-level Note, an inline discussion, and approval are separate objects. Most board cards are views over issue attributes, and review has no atomic object, so preserve every native identity and effect separately.
Before any write:
- Run
glab auth statusand confirm the intended identity on the intended host: a stale env token silently acts as the wrong user. - Read
/metadataand the actual resource/schema when version or tier changes the operation. - Re-fetch the three diff SHAs after every push and verify each inline comment came back as a
DiffNote, never trust a 201 alone.
Requirements
glabonPATH, authenticated to the target host. First-time machine setup (install,glab auth login, protocol, clone, verify) is in references/first-time-setup.md; token types and exact per-operation scopes are in references/auth.md.- A token with
read_apifor read-only work orapi(pluswrite_repositoryfor git ops) for any write, carried by an identity with >= Developer role on the project (or the MR author). jqfor reading API responses (verifying note type, merge status, discussion ids).
Essentials
- Manage workflow tickets: create/reconcile project issues; update metadata without losing unrelated values; preserve project + IID + global ID + URL; and manage native links or work-item-only widgets explicitly. See references/issues.md.
- Manage issue-board kanban: resolve board/list type and IDs, translate card moves into minimal issue-attribute changes, use native Status only on supporting hosts, and keep vertical ranking separate. See references/boards.md.
- Publish durable issue/MR context: use append-only top-level Notes with canonical ID/version/digest markers, exact retry reconciliation, successor links, and native note references. See references/context-notes.md.
- MR object & branch flags:
glab mr create --source-branch <b> --target-branch main(-sdefaults to current branch,-bto project default); glab addresses MRs by per-projectiid. See references/create.md. - glab does not push:
git push -u origin HEADfirst (seegit-guide's push reference), then create; or pass--push. Always pass--yesin scripts or create hangs on the confirmation prompt. See references/create.md. - Summary note: the review's prose is a plain non-anchored note:
glab mr note <iid> -m "## Summary..."(POST .../notes). See references/review-post.md. - Inline comment: a position-anchored discussion:
glab mr note create <iid> --file path --line N -m "...", or rawPOST .../discussionswith a position object carrying three SHAs and conditional line keys. Realizescode-review-guide's findings on GitLab. See references/review-post.md. - Blocking: there is no REST
REQUEST_CHANGES; gate by withholding/approveagainst approval rules or by leaving resolvable threads unresolved withonly_allow_merge_if_all_discussions_are_resolved. See references/review-post.md. - Resolve a thread:
PUT .../discussions/:discussion_id?resolved=true(REST is enough on GitLab; no GraphQL needed); match by discussionid, never by line. See references/review-resolve.md. - Deep-link: MR notes carry no
web_url; build<mr_url>#note_<note_id>from the returned noteid. See references/review-post.md. - Auth & host:
apiscope for writes (read_apireads),GITLAB_HOSTtargets self-managed,GITLAB_TOKENis the general auth token. See references/auth.md. - Enforce with native automation: use pinned, typed, tested CI/CD components, pipeline execution policies, compliance frameworks, protected environments, least-privilege credentials, and provider-native evidence. See references/automation-and-enforcement.md.
- Transact every governance change: discover, preview exact group/project mutations and authority, authorize, apply idempotently against observed revisions, verify with allow and deny probes, roll back, and own the drift. See references/onboarding.md.
- Preserve provider conformance: map tickets, boards, changes, reviews, context, and evidence to separate native identities while exposing tier/version and concurrency limits. See references/provider-conformance.md.
What ships with it
14 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.
- eval-queries.json 5.2 KB
- evals.json 6.9 KB
- references/auth.md 3.3 KB
- references/automation-and-enforcement.md 9.2 KB
- references/boards.md 4.5 KB
- references/context-notes.md 3.4 KB
- references/create.md 3.7 KB
- references/first-time-setup.md 2.6 KB
- references/issues.md 4.6 KB
- references/onboarding.md 3.7 KB
- references/provider-conformance.md 3.6 KB
- references/review-post.md 4.0 KB
- references/review-resolve.md 2.9 KB
- SOURCES.md 8.2 KB
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 · 136 lines · 109 tokens per session scan A 6ac0cdef9cc2
gitlab-guide is a skill published in the GitHub repository xonovex/platform (5 stars, last pushed 2d ago), licensed MIT. It adds 109 tokens to every session and 3,173 once invoked, about $0.0005 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 2026-09-08.
Other skills, from other repositories
work
Handle issue/PR work items with worktree isolation. Triggered by "/work [issuelink|description]" to start focused work.
github-operations
GitHub CLI operations for issues, PRs, milestones, and Projects v2. Covers gh commands, REST API patterns, and automation scripts. Use when managing GitHub issues, PRs, milestones, or Projects with gh.
do
Work an increment task by task through the ledger: task next, claim, implement, commit, task done with evidence. Use when saying "implement", "start working", or "continue increment".
creating-issues-and-pull-requests
Use when creating GitHub pull requests or issues with template compliance. Triggers: 'create a PR', 'open a pull request', 'file an issue', 'create issue'. Also invoked by finishing-a-development-branch. NOT for: deciding whether to merge or PR (use finishing-a-development-branch).
version-bump
Automate version bumping in npm monorepos with CHANGELOG promotion and PR automation.
4x-autopilot
An owner-only automation mode for the 4x feature pipeline, which manages tracked development tasks from selection through completion and merging. It runs one feature at a time and reads project settings to determine the repositories and workflow.