experimental-agent
Experimental tool gated behind experimental-agent flag (disabled)
This record as markdown: /tools/frontmcp/experimental-agent.md
What experimental-agent does on Frontmcp
AI agents invoke experimental-agent to trigger actions in Frontmcp. 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.
Why experimental-agent is rated High
The name 'experimental-agent' strongly implies autonomous agent execution capabilities, which would classify as Execute. However, the tool is described as disabled, and the description is largely uninformative about what it actually does. The 'experimental' label and 'agent' terminology suggest it could trigger complex or unpredictable operations.
From the tool's definition 'experimental-agent' and 'gated behind experimental-agent flag (disabled)'
Attacks that exploit this kind of access
The rule that runs experimental-agent safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Frontmcp, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For experimental-agent, this is the rule to start with:
experimental-agent 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.
The button opens the PolicyLayer dashboard: create your workspace, connect Frontmcp, apply this rule, and every experimental-agent call is checked against it from then on.
Questions about experimental-agent
Experimental tool gated behind experimental-agent flag (disabled). It is categorised as a Execute tool in the Frontmcp MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Front MCP server in PolicyLayer and add a rule for experimental-agent: 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 Frontmcp. Nothing to install.
experimental-agent 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 experimental-agent 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 experimental-agent. 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.
experimental-agent is provided by the Front MCP server (agentfront/frontmcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Front, and thousands of servers like it.
Across the catalogue