AI agents use gmail_draft to create or update resources in Access — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Access environment.
The tool creates or modifies data (a Gmail draft message) in a reversible manner. Drafts can be edited, deleted, or discarded without sending. While it could enable social engineering or unauthorized message composition if an AI agent has unsupervised access to this credential store, the action itself is not destructive or financial.
From the tool's definition Tool description states 'Create a Gmail draft' and 'stage a reply'—these are write operations that create new data in Gmail's draft folder without irreversible deletion or financial impact.
Attacks that exploit this kind of access
Create a Gmail draft without sending it. Use this when a message needs review before sending, or to stage a reply. The draft appears in the user. It is categorised as a Write tool in the Access MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Access MCP server in PolicyLayer and add a rule for gmail_draft: 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 Access. Nothing to install.
gmail_draft 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 gmail_draft 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 gmail_draft. 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.
gmail_draft is provided by the Access MCP server (scottpedia0/access). 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 →