EVM-only: finalize an already-prepared transaction for signing by pinning the nonce, EIP-1559 fees (maxFeePerGas, maxPriorityFeePerGas), and gas limit server-side, then computing the EIP-1559 pre-sign RLP hash Ledger will display in blind-sign mode. Returns a LEDGER BLIND-SIGN HASH content block ...
AI agents use preview_send to create or update resources in VaultPilot MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your VaultPilot MCP environment.
| Parameter | Type | Required | Description |
|---|---|---|---|
handle | string | Yes | Opaque handle returned by a prepare_* tool. preview_send fetches the current nonce + EIP-1559 fees + gas limit, stashes them against the handle, computes the EI |
refresh | boolean | — | Set to true to re-pin nonce/fees/gas (e.g. after the user paused for minutes and wants fresh fees). Default is false: the existing pin and its pre-sign hash are |
Parameters from the server's own tool schema.
An AI agent can call preview_send faster than any human can review — one bad instruction and it creates or modifies resources in VaultPilot MCP by the hundred, each call as confident as the last.
Attacks that exploit this kind of access
EVM-only: finalize an already-prepared transaction for signing by pinning the nonce, EIP-1559 fees (maxFeePerGas, maxPriorityFeePerGas), and gas limit server-side, then computing the EIP-1559 pre-sign RLP hash Ledger will display in blind-sign mode. Returns a LEDGER BLIND-SIGN HASH content block the user reads BEFORE you call send_transaction — the Ledger device prompt blocks the MCP tool call, so the hash must be surfaced now, not after. The pinned tuple is stashed against the handle and forwarded verbatim on send_transaction so the on-device hash is deterministic. If gas conditions drift while the user reviews, call preview_send again on the same handle to refresh the pin (overwrites the prior one). send_transaction will throw a clear error if called without a prior preview_send. Not applicable to TRON handles (USB HID signing flow, no WalletConnect). For Solana handles use preview_solana_send — it pins a fresh blockhash instead of nonce + EIP-1559 fees. It is categorised as a Write tool in the VaultPilot MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
preview_send accepts 2 parameters: handle, refresh. Required: handle. The full parameter table on this page comes from the server's own tool schema.
Register the VaultPilot MCP server in PolicyLayer and add a rule for preview_send: 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 VaultPilot MCP. Nothing to install.
preview_send is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the preview_send 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 preview_send. 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.
preview_send is provided by the VaultPilot MCP server (vaultpilot-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.