AI agents use update_architecture to create or update resources in CoordMCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your CoordMCP environment.
This tool modifies shared architecture state that multiple AI agents rely on for decision-making and consistency. Unlike mere data creation, 'update' suggests changing existing architectural decisions that could affect downstream work. The high severity reflects that corrupted or incorrect architecture updates could misdirect multiple agents, leading to wasted effort or inconsistent implementations.
From the tool's definition Tool name is 'update_architecture' in a coordination server that manages shared project state; the name indicates modification of architecture data that other agents depend on. Description is empty, limiting evidence.
Attacks that exploit this kind of access
update_architecture. It is categorised as a Write tool in the CoordMCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Coord MCP server in PolicyLayer and add a rule for update_architecture: 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 CoordMCP. Nothing to install.
update_architecture 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 update_architecture 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 update_architecture. 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.
update_architecture is provided by the Coord MCP server (siddiquesahabaj/coordmcp). 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 →