AI agents use date_add to create or update resources in Basic — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Basic environment.
This is a Write operation because it transforms data (a date) into a new state. However, the severity is low because: (1) the modification is deterministic and reversible (subtraction undoes addition), (2) no external state is altered—the result is typically used for calculation or display rather than persisted automatically, and (3) the blast radius of misuse is minimal (e.g., incorrect date math in an AI response,…
From the tool's definition The tool 'date_add' modifies a date by adding or subtracting time ('Add or subtract time from a date'). This creates a new/modified datetime value rather than retrieving existing data.
Attacks that exploit this kind of access
Add or subtract time from a date. Use negative amounts to subtract. It is categorised as a Write tool in the Basic MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Basic MCP server in PolicyLayer and add a rule for date_add: 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 Basic. Nothing to install.
date_add 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 date_add 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 date_add. 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.
date_add is provided by the Basic MCP server (msilverblatt/basic-mcp). 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 →