kv2-read
AI agents call kv2-read to retrieve information from Vault MCP Server (mschuchard) without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves secrets from Vault's KV v2 backend without modifying them. While categorized as Read (no side effects), severity is elevated to high because: (1) KV v2 typically stores highly sensitive secrets like database credentials, API keys, and certificates; (2) unauthorized read access could expose credentials used across infrastructure; (3) misuse by an AI agent could leak secrets to logs, external APIs,…
From the tool's definition Tool name 'kv2-read' indicates reading from KV v2 secret engine. Server description states it provides tools for 'interacting with HashiCorp Vault to manage... secret engines like KV v2'.
Attacks that exploit this kind of access
kv2-read. It is categorised as a Read tool in the Vault MCP Server (mschuchard) MCP Server, which means it retrieves data without modifying state.
Register the Vault MCP Server (mschuchard) MCP server in PolicyLayer and add a rule for kv2-read: 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 Vault MCP Server (mschuchard). Nothing to install.
kv2-read is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the kv2-read 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 kv2-read. 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.
kv2-read is provided by the Vault MCP Server (mschuchard) MCP server (mschuchard/vault-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 →