Skip to main content
Glama

Access

gitlab_access
Destructive

Provision, list, rotate, and revoke GitLab access tokens, deploy tokens, deploy keys, and manage access requests and invitations for projects and groups.

Instructions

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

Manage GitLab access credentials: access tokens (project/group/personal), deploy tokens, deploy keys, access requests, and invitations. Revoke/delete actions are destructive and irreversible. When to use: provision and audit who/what can access a project or group; rotate (not revoke+create) to roll a token without invalidating CI configurations. NOT for: SSH/GPG keys or impersonation tokens (use gitlab_user), PAT creation (use gitlab_user create_personal_access_token / create_current_user_pat — gitlab_access exposes token_personal_* for list/get/rotate/revoke only), instance admin operations (use gitlab_admin), project membership/permissions (use gitlab_project member_*), 2FA/MFA flows.

Returns:

  • token_list / deploy_token_list / deploy_key_list_* / request_list_* / invite_list_*: arrays with pagination.

  • token_get / tokencreate / tokenrotate / deploy_token_get / deploy_token_create_* / deploy_key_get / deploy_key_add / deploy_key_update / deploy_key_enable / approve_* / request_*: token / key / request / invitation object. Create / rotate include the cleartext token only ONCE — store it securely; subsequent reads return only the metadata.

  • token_revoke / deploy_token_delete / deploy_key_delete / deny_* : {success, message}. Errors: 401/403 (hint: requires Maintainer+ to manage project tokens, Owner for group, admin for instance / deploy_token_list_all / deploy_key_list_all / deploy_key_add_instance), 404 (hint: token_id and deploy_key_id are scoped to the project/group), 400 (hint: scopes must be a subset of {api, read_api, read_repository, write_repository, read_registry, write_registry}; expires_at must be a future ISO date).

Param conventions: * = required. List actions accept page, per_page. Token actions scope to project_id* or group_id*. Deploy token/key delete and token revoke are irreversible.

Access tokens (token_*) — project, group, and personal scopes. Rotate generates a new token and invalidates the old one:

  • token_project_list / token_group_list: project_id* or group_id*

  • token_project_get / token_group_get: project_id* or group_id*, token_id*

  • token_project_create / token_group_create: project_id* or group_id*, name*, scopes*, expires_at, access_level

  • token_project_rotate / token_group_rotate: project_id* or group_id*, token_id*, expires_at

  • token_project_rotate_self / token_group_rotate_self: project_id* or group_id*, expires_at

  • token_project_revoke / token_group_revoke: project_id* or group_id*, token_id*

  • token_personal_list: user_id

  • token_personal_get: token_id*

  • token_personal_rotate: token_id*, expires_at

  • token_personal_rotate_self: expires_at

  • token_personal_revoke: token_id*

  • token_personal_revoke_self: (no params)

Deploy tokens (deploy_token_*) — scoped to project or group, used for CI/CD registry access:

  • deploy_token_list_all: (admin only)

  • deploy_token_list_project / deploy_token_list_group: project_id* or group_id*

  • deploy_token_get_project / deploy_token_get_group: project_id* or group_id*, deploy_token_id*

  • deploy_token_create_project / deploy_token_create_group: project_id* or group_id*, name*, scopes*, expires_at

  • deploy_token_delete_project / deploy_token_delete_group: project_id* or group_id*, deploy_token_id*

Deploy keys (deploy_key_*) — SSH keys for read/write repo access without a user account. For deploy-key wording, add/create maps to deploy_key_add, fetch/get maps to deploy_key_get, update maps to deploy_key_update, and delete/remove maps to deploy_key_delete.

  • deploy_key_list_project / deploy_key_list_user_project: project_id*

  • deploy_key_list_all: (admin only)

  • deploy_key_get: project_id*, deploy_key_id*. If a workflow says add/create, then fetch/get, then update/delete, call deploy_key_get with the id returned by deploy_key_add before updating.

  • deploy_key_add: project_id*, title*, key*, can_push

  • deploy_key_update: project_id*, deploy_key_id*, title, can_push

  • deploy_key_delete: project_id*, deploy_key_id*

  • deploy_key_enable: project_id*, deploy_key_id*

  • deploy_key_add_instance: title*, key*

Access requests (request_, approve_, deny_*):

  • request_list_project / request_list_group: project_id* or group_id*

  • request_project / request_group: project_id* or group_id*

  • approve_project / approve_group: project_id* or group_id*, user_id*, access_level

  • deny_project / deny_group: project_id* or group_id*, user_id*

Invitations (invite_*):

  • invite_list_project / invite_list_group: project_id* or group_id*

  • invite_project / invite_group: project_id* or group_id*, email*, access_level*, expires_at

See also: gitlab_user (SSH/GPG keys, user PATs), gitlab_admin (instance admin), gitlab_project (project settings)

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?

Beyond annotations (destructiveHint, openWorldHint), the description adds that revoke/delete are irreversible, rotate invalidates old token, cleartext token is returned only once, and error hints (401/403/404/400). No contradiction observed.

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 clear sections (when to use, returns, errors, param conventions, action groups). Some repetition could be trimmed, but the organization helps parseability.

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 complexity (many actions, nested params, return types), the description covers return values, pagination, error patterns, cleartext token disclosure, and per-action parameter lists. The output schema is implicitly described, making it complete.

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 extensive per-action parameter details (required/optional, conventions like project_id*), clarifies scoping, and mentions pagination. The generic params object is fully elaborated.

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 access credentials with a specific verb 'Manage' and resource 'access credentials'. It distinguishes from siblings by explicitly listing what it is NOT for (SSH/GPG keys, PAT creation, etc.) and referencing sibling tools.

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?

Provides explicit when to use (provision/audit/rotate) and when not to use with clear alternatives (gitlab_user, gitlab_admin, gitlab_project). Also offers rotation guidance to avoid breaking CI.

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