AI agents use tool_upload_submission to create or update resources in Gradescope MCP Server — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Gradescope MCP Server environment.
Uploading a submission creates or modifies data in a course management system reversibly (submissions can be resubmitted or removed by normal course operations). This is a Write operation rather than Execute because it accepts a predefined submission object rather than executing arbitrary code. The severity is high because misuse could inject unauthorized submissions, affect grades, or compromise academic integrity.
From the tool's definition Tool name 'upload_submission' indicates creating/adding submission data to Gradescope. The server's stated purpose includes 'assignment management' and 'grading workflows', and this tool fits the pattern of submission intake operations.
Documented attack patterns abuse exactly the kind of access tool_upload_submission gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Gradescope MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for tool_upload_submission:
{
"version": "1",
"default": "deny",
"tools": {
"tool_upload_submission": {
"limits": [
{
"counter": "tool_upload_submission_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}tool_upload_submission stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
tool_upload_submission. It is categorised as a Write tool in the Gradescope MCP Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Gradescope MCP Server MCP server in PolicyLayer and add a rule for tool_upload_submission: 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 Gradescope MCP Server. Nothing to install.
tool_upload_submission 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 tool_upload_submission 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 tool_upload_submission. 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.
tool_upload_submission is provided by the Gradescope MCP Server MCP server (yuanpeng-li/gradescope-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Gradescope MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
37 Gradescope MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.