Skip to main content
Glama
jmrplens

Gitlab MCP Server

by jmrplens
OverviewSchemaRelated ServersScoreDiscussions
Go
Hybrid

Job

gitlab_job
Destructive

Manage GitLab CI/CD jobs: retrieve logs and artifacts, retry or cancel jobs, play manual jobs, and configure job token trust boundaries.

Instructions

Use {"action":"artifacts","params":{...}}; only top-level keys are action and params. Action params schema: gitlab://schema/meta/gitlab_job/.

Manage GitLab CI/CD jobs and the CI/CD job token scope: lifecycle, manual play, log/artifact retrieval, and inbound trust boundaries. Erase/delete actions are destructive. When to use: job details, logs, artifacts, retry/cancel jobs, job token scope. NOT for: pipeline-level operations (use gitlab_pipeline).

Behavior:

  • Idempotent reads: list / list_project / get / trace / artifacts / download_artifacts / download_single_artifact / download_single_artifact_by_ref / list_bridges / token_scope_get / token_scope_list_inbound / token_scope_list_groups.

  • retry starts a NEW job run on every call (NON-idempotent — returns a fresh job_id). play activates an existing manual job that has not yet run (same job_id; only manual jobs with rules.when=manual are eligible) and may pass new variables. cancel is idempotent (no-op once final). keep_artifacts / token_scope_patch / token_scope_add_project / token_scope_add_group are idempotent.

  • Side effects: retry / play queue runners, consume CI minutes, and may trigger downstream pipelines and notifications. trace returns up to 100KB of log; download_artifacts streams up to 1MB inline (base64).

  • Destructive: erase clears the job log and artifacts in place (irreversible); delete_artifacts removes a single job's artifacts; delete_project_artifacts wipes ALL artifacts across the project (irreversible). token_scope_remove_* tightens trust boundaries and may break running pipelines.

Param conventions: * = required. All job actions need project_id*. List actions accept page, per_page.

Jobs:

  • list: project_id*, pipeline_id*, scope

  • list_project: project_id*, scope, include_retried

  • get: project_id*, job_id*

  • trace: project_id*, job_id*. Returns job log (truncated to 100KB).

  • cancel / retry / erase / keep_artifacts: project_id*, job_id*

  • play: project_id*, job_id*, variables (array of {key, value, variable_type})

  • wait: project_id*, job_id*, interval_seconds (5-60, default 10), timeout_seconds (1-3600, default 300), fail_on_error (default true)

  • list_bridges: project_id*, pipeline_id*, scope

  • delete_artifacts: project_id*, job_id*

  • delete_project_artifacts: project_id*. Deletes ALL artifacts across project.

Artifact downloads (base64, max 1MB):

  • artifacts: project_id*, job_id* — download the whole artifact archive from a known numeric job ID.

  • download_artifacts: project_id*, ref_name*, job* — download the whole artifact archive by ref_name and job NAME (string). Never use with job_id.

  • download_single_artifact: project_id*, job_id*, artifact_path* — use when the prompt gives a numeric job ID and one artifact file path such as coverage/report.xml. This is the single-file-by-job-id action.

  • download_single_artifact_by_ref: project_id*, ref_name*, artifact_path*, job* — use when the prompt gives ref_name plus job NAME and one artifact file path. Never use with job_id.

Job token scope:

  • token_scope_get / token_scope_patch: project_id*. Patch params: enabled.

  • token_scope_list_inbound: project_id*

  • token_scope_add_project / token_scope_remove_project: project_id*, target_project_id*

  • token_scope_list_groups: project_id*

  • token_scope_add_group / token_scope_remove_group: project_id*, target_group_id*

See also: gitlab_pipeline, gitlab_repository

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
actionYesAction to perform. Pick exactly one of the values in `enum`. Each action expects its own `params` object — see the tool description for the per-action parameter list.
paramsNoAction-specific parameters as a JSON object. Required and optional fields differ per action. This envelope schema stays broad; runtime validation applies the chosen action's schema after reserved meta keys like `confirm` are stripped. For the JSON Schema of a specific action's `params`, read the MCP resource `gitlab://schema/meta/{tool}/{action}` (replace placeholders with the tool name and the chosen action).

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
next_stepsNoOptional. Suggested follow-up actions or tool calls for the LLM, contextual to the result.
paginationNoPresent on list actions. Use `has_more` and `next_page` to paginate through results.

Tool Definition Quality

A4.9/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description details idempotent vs non-idempotent actions, side effects (CI minutes consumption, downstream triggers), destructive operations (erase, delete_artifacts), and size limits (trace 100KB, artifacts 1MB). Annotations only indicate destructive hint; the description provides full context.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is very detailed but well-organized with sections (Behavior, Jobs, Artifact downloads, etc.). It front-loads the action format. Minor verbosity in listing all actions, but overall structured.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool has 25 actions and nested parameters, the description covers all actions, their parameters, behavior, side effects, and distinguishes between similar actions (e.g., artifact download variants). It is self-contained and comprehensive.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema is generic with action and params, but the description lists all per-action parameters with required status and optional fields. It clarifies conventions like 'project_id*' and artifact download variants.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses specific verbs like 'manage', 'retrieve', and lists distinct actions (list, get, retry, cancel, etc.). It clearly distinguishes from sibling tools like gitlab_pipeline and gitlab_repository.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It explicitly states when to use the tool ('job details, logs, artifacts, retry/cancel jobs, job token scope') and when not to ('NOT for: pipeline-level operations (use gitlab_pipeline)'). It also names alternative tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/jmrplens/gitlab-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server