AI agents use add_document to create or update resources in Local Rag — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Local Rag environment.
This tool creates or modifies the indexed knowledge base by adding documents. It is reversible (documents can be removed), so it falls under Write rather than Destructive. The severity is medium because misuse could pollute the knowledge base with incorrect or malicious content, but the impact is limited to the local RAG system and can be corrected by removing the added documents.
From the tool's definition Tool name 'add_document' and description indicate it adds files to a RAG index ('ファイル(PDF / txt / md)を RAG インデックスに追加します' = 'Add files (PDF/txt/md) to RAG index'). This is a create/modify operation that writes data to the knowledge base.
Attacks that exploit this kind of access
ファイル(PDF / txt / md)を RAG インデックスに追加します。. It is categorised as a Write tool in the Local Rag MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Local Rag MCP server in PolicyLayer and add a rule for add_document: 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 Local Rag. Nothing to install.
add_document 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_document 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_document. 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_document is provided by the Local Rag MCP server (n-irei/local-rag-mcp-server). 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 →