Send a previously created draft.
AI agents use keyid_send_draft to create or update resources in KeyID Agent Kit — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your KeyID Agent Kit environment.
This tool sends an email draft to recipients. While sending an email is technically irreversible in the sense that the recipient receives it, the operation itself is a standard Write action (creating/transmitting a message). It is not Destructive because the draft and original data are not permanently erased.
From the tool's definition Tool name 'keyid_send_draft' and description 'Send a previously created draft' indicate sending an email message, which is a reversible write operation that modifies the state of the email system and affects external recipients.
Attacks that exploit this kind of access
Send a previously created draft. It is categorised as a Write tool in the KeyID Agent Kit MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the KeyID Agent Kit MCP server in PolicyLayer and add a rule for keyid_send_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 KeyID Agent Kit. Nothing to install.
keyid_send_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 keyid_send_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 keyid_send_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.
keyid_send_draft is provided by the KeyID Agent Kit MCP server (keyid-ai/agent-kit). 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 →