Skip to main content
Glama

Runner

gitlab_runner
Destructive

Manage GitLab CI/CD runners: create, update, delete, register, and control runner tokens and scopes for instance, group, and project runners.

Instructions

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

Manage GitLab CI/CD runners (instance, group, project) and runner controllers (admin, experimental): CRUD, registration tokens, and job assignments. Remove/delete/revoke and reset_token actions are destructive — revoking the registration token only invalidates future registrations; already-registered runners keep operating using their existing runner authentication tokens. Valid actions: controller_create, controller_delete, controller_get, controller_list, controller_scope_add_instance, controller_scope_add_runner, controller_scope_list, controller_scope_remove_instance, controller_scope_remove_runner, controller_token_create, controller_token_get, controller_token_list, controller_token_revoke, controller_token_rotate, controller_update, delete_by_token, delete_registered, disable_project, enable_project, get, jobs, list, list_all, list_group, list_managers, list_project, register, remove, reset_group_reg_token, reset_instance_reg_token, reset_project_reg_token, reset_token, update, verify

When to use: register or pause runners, change runner tags / access_level / maximum_timeout, attach or detach runners from a project / group, rotate registration tokens, drive runner controllers (CRUD + scopes + tokens) for admins. NOT for: pipeline runs (use gitlab_pipeline), job logs / retry / play (use gitlab_job), CI variables (use gitlab_ci_variable), CI lint or templates (use gitlab_template), self-hosted GitLab Runner installation (out of scope — install via the GitLab Runner CLI).

Returns:

  • list / list_all / list_project / list_group / list_managers / jobs / controller_list / controller_scope_list / controller_token_list: arrays with pagination {page, per_page, total, next_page}.

  • get / update / register / reset_token / enable_project / controller_get / controller_create / controller_update / controller_scope_add / controller_token_get / controller_token_create / controller_token_rotate: runner / controller / token object. register and reset_*_token / controller_token_create / controller_token_rotate include the cleartext token only ONCE — store it securely.

  • verify / remove / delete_registered / delete_by_token / disable_project / controller_delete / controller_scope_remove_* / controller_token_revoke: {success, message}. Errors: 401/403 (hint: list_all / register with admin token / runner controller actions require admin), 404 (hint: runner_id and controller_id are global, project / group context only filters), 400 (hint: access_level ∈ not_protected / ref_protected; tag_list is a comma-separated string; deprecated reset_*_reg_token endpoints — prefer controller_token_create).

Param conventions: * = required. List actions accept page, per_page. Runner IDs are integers.

Runner CRUD:

  • list: type, status, paused, tag_list (comma-separated)

  • list_all: (admin) type, status, paused, tag_list

  • get / remove: runner_id*

  • update: runner_id*, description, paused, tag_list, run_untagged, locked, access_level, maximum_timeout, maintenance_note

  • jobs: runner_id*, status (running/success/failed/canceled), order_by, sort, page, per_page

Project/Group runners:

  • list_project: project_id*, type, status, tag_list

  • enable_project: project_id*, runner_id*

  • disable_project: project_id*, runner_id*

  • list_group: group_id*, type, status, tag_list

Registration and tokens:

  • register: token*, description, paused, locked, run_untagged, tag_list, access_level, maximum_timeout, maintenance_note

  • delete_registered: runner_id*

  • delete_by_token: token*

  • verify: token*

  • reset_token: runner_id*

  • reset_instance_reg_token: (deprecated, no params)

  • reset_group_reg_token: group_id* (deprecated)

  • reset_project_reg_token: project_id* (deprecated)

  • list_managers: runner_id*

Runner controllers (admin, experimental):

  • controller_list

  • controller_get / controller_delete: controller_id*

  • controller_create: description, state (enabled/disabled/dry_run)

  • controller_update: controller_id*, description, state

Controller scopes:

  • controller_scope_list / controller_scope_add_instance / controller_scope_remove_instance: controller_id*

  • controller_scope_add_runner / controller_scope_remove_runner: controller_id*, runner_id*

Controller tokens:

  • controller_token_list: controller_id*

  • controller_token_get / controller_token_rotate / controller_token_revoke: controller_id*, token_id*

  • controller_token_create: controller_id*, description

See also: gitlab_pipeline, gitlab_job

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.
Behavior5/5

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

The description notes that remove/delete/revoke and reset_token actions are destructive, warns that registration tokens are shown only once, and provides error hints (401/403, 404, 400). Annotations include destructiveHint: true, consistent with the description. No contradiction.

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 lengthy but well-structured with sections (valid actions, returns, errors, param conventions) and is front-loaded with the critical action/params structure. Some redundancy exists, but it remains clear.

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?

Covers return types for every action (arrays, objects, success messages), pagination details, error hints, deprecation notes, and even references a schema resource for per-action validation. The presence of an output schema helps, but the description adds significant context.

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?

Despite 100% schema coverage, the description adds per-action parameter lists, conventions (e.g., required fields, pagination), and specific parameter details (e.g., access_level values, tag_list format). This greatly aids correct invocation.

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 clearly states the tool manages GitLab CI/CD runners and runner controllers with CRUD, registration tokens, and job assignments. It distinguishes from sibling tools by listing what it is NOT for (e.g., gitlab_pipeline, gitlab_job).

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?

A dedicated 'When to use' section lists appropriate use cases, and a 'NOT for' section explicitly names alternative tools, providing clear guidance on when not to use this tool.

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