BlurHook MCP
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| BLURHOOK_POLICY | No | Supplies inline policy JSON. | |
| BLURHOOK_RPC_URLS | No | Chooses read RPCs. | |
| BLURHOOK_POOLS_FILE | No | Adds pool definitions. | |
| BLURHOOK_LEDGER_FILE | No | Persists daily usage across restarts. Without this, the default in-memory ledger resets when the process restarts. | |
| BLURHOOK_POLICY_FILE | No | Selects a policy JSON file. | |
| BLURHOOK_DISCOVER_BLOCKS | No | Enables a bounded startup discovery scan. Default discovery is off. | off |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| blurhook_deploymentA | Read BlurHook contract addresses and privacy scope on Robinhood Chain. This tool does not construct private proofs or authorize deposits. |
| quoteA | Best route for a swap on Robinhood Chain across Uniswap v3 (4 fee tiers), v4 and hooked v4 pools, including splits and two-hop routes via USDG or ETH. Returns amounts, minimum received, price impact, pinned-code and hook checks, warnings, and a preview of the spending-policy verdict. Never returns calldata. |
| build_swapA | Quote, enforce the spending policy (allowlist, per-trade cap, daily cap, max slippage, max price impact, pinned contracts and hooks), simulate from the agent wallet, and return Universal Router calldata for that wallet to sign. ERC-20 input without Permit2 allowances returns the approval calldata first. Refusals list every rule that failed. The trade's USD size counts toward the daily cap once calldata is returned. |
| simulateA | Run a transaction with eth_call from a wallet and estimate its gas, without sending it. Returns ok, gasEstimate and the revert reason when it fails. fundEth gives the wallet enough ETH through a state override for a dry run. |
| list_hook_poolsA | Hooks in the server's pool list ranked by swaps, with pool counts, tokens and recorded codehash pins. Pass a hook to list its pools, or a token to filter. |
| hook_passportA | What a Uniswap v4 hook can do, read from public on-chain data: permission flags from its address, proxy and owner, Sourcify verification, admin-style functions in its bytecode, pools in the list, and a rule-based risk score with every rule that fired. |
| token_priceA | USD price per whole token from live Uniswap v3/v4 quotes, directly against USDG or via ETH (USDG counts as 1 USD). Up to 30 tokens. |
| stock_tokensA | Robinhood Stock Tokens on Robinhood Chain, listed only when the contract's runtime code exactly matches the official Stock Token proxy (copycats with the same symbol are never listed), with whether the spending policy allows trading each. |
| spending_policyA | The spending policy this server enforces before returning calldata, the USD spent in the rolling 24 h window and what is left of the daily cap. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
Most tools target distinct resources or workflow stages: quote, build_swap, simulate, spending_policy, hook discovery, hook risk, pricing, and stock-token listing. The main overlap is that build_swap internally performs quoting and simulation, and quote previews the spending-policy verdict, but the descriptions make the boundaries reasonably clear.
All names use snake_case, which is consistent, but the convention mixes noun phrases (spending_policy, hook_passport, token_price, stock_tokens) with verb or verb_noun forms (quote, simulate, build_swap, list_hook_pools). It remains readable, but an agent cannot reliably infer a tool's action from a predictable naming pattern.
Nine tools is well within the appropriate range and each appears to serve a distinct role in the swap workflow, policy enforcement, hook safety, and token discovery. The set does not feel bloated or artificially thin.
The core lifecycle is covered: policy inspection, quoting, calldata construction, simulation, hook discovery and risk analysis, pricing, and stock-token validation. Minor gaps remain around direct allowance/balance inspection and generic pool or token discovery outside the hook and stock-token lists, but these are workable around the documented flow.