Call this tool when you need to understand how to run experiments and evaluations in LangSmith.
AI agents invoke run_experiment to trigger actions in LangSmith MCP Server. 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.
This tool executes experiments and evaluations in LangSmith, which constitutes triggering external operations. The effects are dependent on the experiment configuration passed as arguments. While not as severe as Destructive (experiments are typically reversible) or Financial, it clearly goes beyond Read/Write categories and into Execute territory.
From the tool's definition 'run experiments and evaluations' indicates the tool triggers external operations whose effects depend on arguments.
Documented attack patterns abuse exactly the kind of access run_experiment gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and LangSmith MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for run_experiment:
{
"version": "1",
"default": "deny",
"tools": {
"run_experiment": {
"limits": [
{
"counter": "run_experiment_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}run_experiment stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Call this tool when you need to understand how to run experiments and evaluations in LangSmith. It is categorised as a Execute tool in the LangSmith MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the LangSmith MCP Server MCP server in PolicyLayer and add a rule for run_experiment: 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 LangSmith MCP Server. Nothing to install.
run_experiment 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 run_experiment 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 run_experiment. 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.
run_experiment is provided by the LangSmith MCP Server MCP server (langchain-ai/langsmith-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from LangSmith MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
15 LangSmith MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.