AI agents use add_file to create or update resources in MCP RAG — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MCP RAG environment.
Adding a file creates or modifies the vector storage and document index reversibly—a file can be removed later via remove_file. This is a Write operation, not Read (which would be search_files or list_files) and not Destructive (no deletion).
From the tool's definition Tool description states 'Add a file to the RAG system' which is a create/write operation that modifies the document store. The sibling tools include destructive operations (remove_file, remove_memory, clear_data) but add_file itself only creates new entries.
Attacks that exploit this kind of access
Add a file to the RAG system for document retrieval. It is categorised as a Write tool in the MCP RAG MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the MCP RAG MCP server in PolicyLayer and add a rule for add_file: 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 RAG. Nothing to install.
add_file 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_file 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_file. 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_file is provided by the MCP RAG MCP server (santis84/mcp-rag). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
add_file is one line of MCP RAG's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →