go_test
AI agents invoke go_test to trigger actions in Mcp Golang. 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.
Running tests executes arbitrary code in the Go test suite. An AI agent misusing this tool could trigger test code with side effects, resource consumption, or network calls. The empty description lowers confidence, but the tool name and server context strongly suggest code execution. Severity is high because executing arbitrary test code can have broad side effects depending on what the tests do.
From the tool's definition Tool name 'go_test' on a server that 'runs Go language tools' including analyze, fix, and test operations. The description is empty, but 'go test' in Go conventionally executes test code.
Attacks that exploit this kind of access
go_test. It is categorised as a Execute tool in the Mcp Golang MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Mcp Golang MCP server in PolicyLayer and add a rule for go_test: 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 Mcp Golang. Nothing to install.
go_test 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 go_test 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 go_test. 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.
go_test is provided by the Mcp Golang MCP server (rethunk-ai/rethunk-mcp-go). 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 →