AI agents use manage_kv_store to create or update resources in Scouter — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Scouter environment.
The term 'manage' applied to a 'kv_store' (key-value store) implies creating, updating, or modifying stored data rather than merely reading it. Within an APM context, this could affect configuration or monitoring state. The vague description reduces confidence, but the management verb and data store type indicate Write-level capability. Not Destructive unless deletion is explicitly supported (not stated).
From the tool's definition Tool name 'manage_kv_store' and description 'Manage Scouter' indicate management operations on a key-value store. HTTP mode context suggests network-based state modification.
Attacks that exploit this kind of access
[HTTP mode only] Manage Scouter. It is categorised as a Write tool in the Scouter MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Scouter MCP server in PolicyLayer and add a rule for manage_kv_store: 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 Scouter. Nothing to install.
manage_kv_store 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 manage_kv_store 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 manage_kv_store. 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.
manage_kv_store is provided by the Scouter MCP server (scouter-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 →