AI agents call get_descriptions to retrieve information from dexMCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves Pokedex flavor text (descriptive/lore data about Pokemon), which is a pure read operation with no capability to modify, delete, or execute code. The data fetched is informational and immutable from the user's perspective. No financial, destructive, or execution-based side effects are possible.
From the tool's definition Tool name 'get_descriptions' and description 'Fetch localized Pokedex flavor text' indicate data retrieval with no side effects. The verb 'Fetch' and context of retrieving read-only Pokemon lore/descriptions confirm this is a query operation.
Attacks that exploit this kind of access
Fetch localized Pokedex flavor text. It is categorised as a Read tool in the dexMCP MCP Server, which means it retrieves data without modifying state.
Register the dex MCP server in PolicyLayer and add a rule for get_descriptions: 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 dexMCP. Nothing to install.
get_descriptions 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 get_descriptions 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 get_descriptions. 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.
get_descriptions is provided by the dex MCP server (rajeevatla/dexmcp). 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 →