AI agents use upload_chat to create or update resources in Magnet — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Magnet environment.
This tool creates a new chat session record in Magnet. It is reversible (the chat could be deleted afterward) and does not execute arbitrary code, delete data irreversibly, or move money. It falls squarely in the Write category.
From the tool's definition Tool description states 'Upload a chat session to Magnet' and 'Returns the created chat', indicating creation of new data. The name 'upload_chat' and verb 'Upload' confirm this is a write operation that creates a new resource.
Attacks that exploit this kind of access
Upload a chat session to Magnet for tracking and analysis. Returns the created chat with a viewUrl to see it in the web UI. It is categorised as a Write tool in the Magnet MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Magnet MCP server in PolicyLayer and add a rule for upload_chat: 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 Magnet. Nothing to install.
upload_chat 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 upload_chat 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 upload_chat. 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.
upload_chat is provided by the Magnet MCP server (@magnet-ai/magnet-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 →