AI agents call fetch_runs to retrieve information from LangSmith MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool appears designed to query and retrieve historical run/trace data from LangSmith for observability purposes. No side effects or data modification is implied by the name or the pattern of sibling tools. This is consistent with Read category tools that retrieve observability metrics and historical data without altering system state.
From the tool's definition Tool name 'fetch_runs' indicates retrieval of run data from LangSmith observability platform. Sibling tools include 'get_*' and 'list_*' operations typical of read-only access patterns. Description is empty, which lowers confidence slightly.
Documented attack patterns abuse exactly the kind of access fetch_runs 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 fetch_runs:
{
"version": "1",
"default": "deny",
"tools": {
"fetch_runs": {}
}
}fetch_runs is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
fetch_runs. It is categorised as a Read tool in the LangSmith MCP Server MCP Server, which means it retrieves data without modifying state.
Register the LangSmith MCP Server MCP server in PolicyLayer and add a rule for fetch_runs: 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.
fetch_runs 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 fetch_runs 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 fetch_runs. 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.
fetch_runs 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.