AI agents call gitlab_get_project to retrieve information from Sage MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves project metadata/details from GitLab. It performs a read-only query operation with no side effects, no data modification, and no irreversible actions. The blast radius of misuse is minimal since it only exposes existing project information that would typically be accessible to users with appropriate permissions. No code execution, financial impact, or destructive capability is present.
From the tool's definition Tool name is 'gitlab_get_project' and description states 'Get detailed information about a project' - both indicate retrieval of information without modification or deletion.
Documented attack patterns abuse exactly the kind of access gitlab_get_project gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Sage MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for gitlab_get_project:
{
"version": "1",
"default": "deny",
"tools": {
"gitlab_get_project": {}
}
}gitlab_get_project is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Get detailed information about a project. It is categorised as a Read tool in the Sage MCP MCP Server, which means it retrieves data without modifying state.
Register the Sage MCP server in PolicyLayer and add a rule for gitlab_get_project: 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 Sage MCP. Nothing to install.
gitlab_get_project is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the gitlab_get_project 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 gitlab_get_project. 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.
gitlab_get_project is provided by the Sage MCP server (sagemcp/sagemcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 359 Sage MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
359 Sage MCP tools catalogued and risk-classified — across an index of 42,500+ MCP servers.