Get the agent
AI agents call get_agent_instructions to retrieve information from Cross-Project MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves information about agent instructions, which is a read-only operation with no side effects. The incomplete description ('Get the agent') limits confidence slightly, but the pattern of sibling tools and the nature of instruction retrieval (typically informational) place this clearly in the Read category.
From the tool's definition Tool name is 'get_agent_instructions' and retrieves agent instructions. The description fragment 'Get the agent' combined with the context of a cross-project code exploration server indicates this retrieves configuration or metadata without modifying anything.
Attacks that exploit this kind of access
Get the agent. It is categorised as a Read tool in the Cross-Project MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Cross-Project MCP Server MCP server in PolicyLayer and add a rule for get_agent_instructions: 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 Cross-Project MCP Server. Nothing to install.
get_agent_instructions 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 get_agent_instructions 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 get_agent_instructions. 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.
get_agent_instructions is provided by the Cross-Project MCP Server MCP server (sweetsrepo/cross-project-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 →