Execute any Cosmos SDK query command. Call list_modules and list_module_subcommands first to discover available options.
AI agents invoke cosmos_query to trigger actions in Manifest MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
The tool explicitly executes arbitrary Cosmos SDK query commands. While 'query' suggests read-only operations, the ability to execute any command means the scope depends on arguments and could extend beyond simple reads. The instruction to discover available options via list_modules further implies broad, dynamic execution capability.
From the tool's definition "Execute any Cosmos SDK query command"
Attacks that exploit this kind of access
Execute any Cosmos SDK query command. Call list_modules and list_module_subcommands first to discover available options. It is categorised as a Execute tool in the Manifest MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Manifest MCP server in PolicyLayer and add a rule for cosmos_query: 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 Manifest MCP. Nothing to install.
cosmos_query is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the cosmos_query 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 cosmos_query. 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.
cosmos_query is provided by the Manifest MCP server (manifest-network/manifest-mcp-mono). 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 →