Upload a file from a sandbox/container path (e.g. /mnt/user-data/, /home/claude/) that the host cannot access. Reads the file and uploads it directly — single tool call, no shell commands needed.
AI agents use upload_from_sandbox to create or update resources in Mcp Upload — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mcp Upload environment.
This tool creates new data (file upload to storage service) with persistent, reversible effects. It is Write rather than Execute because it does not run arbitrary code or commands—it specifically reads and uploads a file.
From the tool's definition Tool description states 'Upload a file' and 'uploads it directly', creating a persistent permalink. Sibling tool 'upload_file' confirms the server's write capability. The mechanism reads and transfers file content to persistent storage.
Attacks that exploit this kind of access
Upload a file from a sandbox/container path (e.g. /mnt/user-data/, /home/claude/) that the host cannot access. Reads the file and uploads it directly — single tool call, no shell commands needed. It is categorised as a Write tool in the Mcp Upload MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mcp Upload MCP server in PolicyLayer and add a rule for upload_from_sandbox: 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 Upload. Nothing to install.
upload_from_sandbox 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_from_sandbox 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_from_sandbox. 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_from_sandbox is provided by the Mcp Upload MCP server (@objekt.sh/mcp-upload). 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 →