AI agents use patch_connection_tags to create or update resources in Nango — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Nango environment.
The tool updates a connection's tag set, which is a reversible data modification operation. It does not irreversibly delete data (not Destructive), does not execute arbitrary code (not Execute), and does not involve financial transactions (not Financial). The advisory to 'fetch and merge first' confirms this is a write operation where overwrites are possible.
From the tool's definition patch_connection_tags: Replace a connection's complete tag set. This modifies existing connection metadata reversibly.
Attacks that exploit this kind of access
Replace a connection's complete tag set. Fetch and merge first when changing one tag. It is categorised as a Write tool in the Nango MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Nango MCP server in PolicyLayer and add a rule for patch_connection_tags: 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 Nango. Nothing to install.
patch_connection_tags 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 patch_connection_tags 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 patch_connection_tags. 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.
patch_connection_tags is provided by the Nango MCP server (levsky22/nango-mcp-server). 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 →