search_project_tool
AI agents call search_project_tool to retrieve information from MCP Gitlab without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Search operations are non-destructive data retrieval. Given the GitLab server context where other tools handle mutations (create_*, cancel_*), this tool almost certainly queries project data without side effects. Low severity due to no capability to modify or destroy data. Confidence reduced slightly due to missing description.
From the tool's definition Tool name 'search_project_tool' indicates a search operation, which retrieves or queries data. The empty description prevents full certainty, but the name strongly suggests a read-only retrieval function consistent with project search operations.
Attacks that exploit this kind of access
search_project_tool. It is categorised as a Read tool in the MCP Gitlab MCP Server, which means it retrieves data without modifying state.
Register the MCP Gitlab MCP server in PolicyLayer and add a rule for search_project_tool: 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.
search_project_tool 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 search_project_tool 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 search_project_tool. 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.
search_project_tool is provided by the MCP Gitlab MCP server (mcp-gitlab-crunchtools). 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 →