Pre-trade gate for agents on Robinhood Chain: signed ALLOW, REVIEW or DENY before a token buy.
Scored across 11 tools
The gate family (gate, gate_batch, gate_facts) and the mandate family (create/cancel_mandate, mandate_status/events/signal) are cleanly separated by resource and intent. The closest pairs are get_receipt vs verify_receipt and mandate_signal vs mandate_events, but the descriptions make the retrieve-vs-authenticate and newest-signal-vs-log distinction explicit enough to avoid misselection.
All names are snake_case and cluster under predictable prefixes (gate*, mandate_*, verify_*, *_receipt), which reads well. Minor deviation: some names are verb_noun (cancel_mandate, get_receipt) while others are noun-only or bare-verb (gate, gate_facts, mandate_signal, mandate_events), so the verb position is not fully uniform.
11 tools is well within the sweet spot for this domain, and each one has a distinct job: gating, batching, fact discovery, mandate lifecycle, polling, and two verification paths. No tool looks like filler.
The core loop is covered end to end: create/cancel a mandate, poll events, fetch and verify signals, gate tokens singly or in batch, inspect facts, and store/verify receipts. The main gap is enumeration and mutation — there is no list_mandates and no way to change a mandate's policy or delivery mode after creation, forcing agents to retain ids and re-create mandates.
Stops checks and events for a mandate, e.g. after selling the token.
Inputs: id (required), secret (required).
For a token you hold: $0.05 per 7 days over x402 (pay via the REST API or the stdio MCP server). Re-checks a token against the policy about every 30 minutes. Changes (decision, owner, implementation, mint/pause/blacklist, supply) are recorded as events you poll with mandate_events; optionally also POSTed to an https webhook. Every event carries a signed signal: exit, watch or hold. Returns an id and a secret: keep the secret.
Inputs: token (required), policy, exit_on, webhook_url.
Call BEFORE buying a token on Robinhood Chain. Returns ALLOW, DENY or REVIEW, every check with the fact it read, and an EIP-712 signed receipt. Missing data never gives ALLOW unless the policy says so.
Inputs: token (required), policy.
Gate up to 10 tokens against one policy in one call, one receipt each. Counts as one Gate call per token.
Inputs: tokens (required), policy.
Every fact a policy rule may use (name, type, meaning), the rule syntax, the presets, and current pricing.
Inputs: no named parameters.
A stored decision receipt by id (rcpt_…).
Inputs: id (required).
Events a mandate recorded since a unix time (0 = all). Use next_since for the following poll.
Inputs: id (required), secret (required), since.
The newest signed signal for a mandate: exit, watch or hold, from its last decision, valid for one hour. Verify it, then your own code decides whether to sell; Actuary never trades.
Inputs: id (required), secret (required).
Current decision, delivery mode, last check and failures of a mandate.
Inputs: id (required), secret (required).
Checks that a receipt was signed by Actuary and that nothing in it changed. Pass the receipt object or its id.
Inputs: receipt (required).
Checks that a mandate signal was signed by Actuary, that nothing in it changed and that it is still fresh. Act only on valid and fresh.
Inputs: signal (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Deterministic pre-trade risk checks for AI agents on Robinhood Chain. Returns allow/warn/deny.
Pre-execution governance for AI agents. Deterministic PASS/FAIL/REVIEW verdicts, replayable proof.
Robinhood Chain checks for agents: coins, devs, ERC-8004 agents, pre-sign check, Stock Token data.
Pre-execution policy gate for consequential agent actions with durable trust receipts.
Add one secure layer between your agents and this server.
