Get files modified in a session
AI agents call get_session_files to retrieve information from Mcp Coordinator without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves information about files that have been modified in a session. This is a read-only operation that queries state without creating, modifying, deleting, or executing anything. The blast radius is minimal — an agent learning which files were touched in a session poses no direct risk. Low severity reflects the informational nature and lack of destructive or state-altering capability.
From the tool's definition Tool name is 'get_session_files' and description states 'Get files modified in a session' — this is a query/retrieval operation with no modification or side effects.
Attacks that exploit this kind of access
Get files modified in a session. It is categorised as a Read tool in the Mcp Coordinator MCP Server, which means it retrieves data without modifying state.
Register the Mcp Coordinator MCP server in PolicyLayer and add a rule for get_session_files: 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 Coordinator. Nothing to install.
get_session_files 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_session_files 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_session_files. 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_session_files is provided by the Mcp Coordinator MCP server (swoofer/mcp-coordinator). 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 →