Open a visible (headed) browser window preserving all current state — cookies, tabs, and localStorage. Use when the user is blocked by CAPTCHAs, bot detection, or complex auth flows that require manual intervention in a headed browser. After the user solves the challenge, call pilot_resume to rec...
AI agents invoke pilot_handoff to trigger actions in Pilot. 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 launches a headed browser window while preserving all session state (cookies, localStorage, tabs), which constitutes triggering an external operation. It doesn't read data, write/modify data directly, or delete anything, but it executes a browser state transition that exposes the live authenticated session visually and enables manual user interaction.
From the tool's definition Open a visible (headed) browser window preserving all current state — cookies, tabs, and localStorage... triggers external operations (opening a browser window with full session state)
Documented attack patterns abuse exactly the kind of access pilot_handoff gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pilot, and nothing reaches the server without passing your rules. This is the rule we recommend for pilot_handoff:
{
"version": "1",
"default": "deny",
"tools": {
"pilot_handoff": {
"limits": [
{
"counter": "pilot_handoff_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}pilot_handoff 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.
Open a visible (headed) browser window preserving all current state — cookies, tabs, and localStorage. Use when the user is blocked by CAPTCHAs, bot detection, or complex auth flows that require manual intervention in a headed browser. After the user solves the challenge, call pilot_resume to reclaim automated control. Parameters: (none) Returns: Confirmation that the browser is now in headed mode with instructions to call pilot_resume when done. Errors: -. It is categorised as a Execute tool in the Pilot MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Pilot MCP server in PolicyLayer and add a rule for pilot_handoff: 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 Pilot. Nothing to install.
pilot_handoff 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 pilot_handoff 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 pilot_handoff. 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.
pilot_handoff is provided by the Pilot MCP server (tacosyhorchata/pilot). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 61 Pilot tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
61 Pilot tools catalogued and risk-classified — across an index of 42,500+ MCP servers.