Add a user to a project with a given access level. dry_run=true by default.
AI agents use add_project_member to create or update resources in Mcp Gitlab — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mcp Gitlab environment.
This tool modifies project access control by creating a new member record, which is reversible (members can be removed). It is Write rather than Execute because it manipulates data structures (membership records) rather than triggering arbitrary operations. The severity is high because granting incorrect access levels could expose sensitive project data or enable unauthorized actions.
From the tool's definition add_project_member: 'Add a user to a project with a given access level' — this creates a new project membership binding with defined permissions.
Attacks that exploit this kind of access
Add a user to a project with a given access level. dry_run=true by default. It is categorised as a Write tool in the Mcp Gitlab MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mcp Gitlab MCP server in PolicyLayer and add a rule for add_project_member: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Mcp Gitlab. Nothing to install.
add_project_member is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the add_project_member rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for add_project_member. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
add_project_member is provided by the Mcp Gitlab MCP server (wanadev/gitlab-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →