Insight Pre-Trade Safety
Decision-input integrity for onchain AI agents: verify oracle prices before signing a trade.
| Endpoint | https://www.oracleinsight.xyz/api/mcp (streamable-http) |
|---|---|
| Registry version | 0.3.1 |
| Now | 16 tools · grade C (69) · pin ts_e259e029d6daf4edc69ea2ba0ddc001a · protocol 2025-06-18 |
| First / last listed | 2026-11-08 01:38 UTC / 2026-11-08 01:38 UTC |
| Grade over time | 2026-11-08: C (69), 16 tools |
Is this your server? Own a server? ToolFit hosted watch re-checks it every 24 h ($0.25 per 30 days) or 6 h ($0.60) and sends you an HMAC-signed webhook when its tools or grade change: POST https://api.youraitools.co/toolfit/v1/watch. To opt out, see About & opt-out.
Tools: get_oracle_price get_price_analytics get_oracle_health search_oracle_registry get_market_quote check_trade_safety begin_trade_evidence prepare_swap prepare_perp_order issue_execution_receipt verify_evidence oracle_watch oracle_watch_window_begin oracle_watch_window_control position_lifecycle_begin position_lifecycle_step
get_oracle_price | Read the current price one oracle provider reports for one asset. Use it for a single fresh or cached provider read (for example when a route quote needs a reference price). Do not use it for cross-provider consensus, historical series, deviation or risk analytics — use get_price_analytics instead. Returns the price, provider, symbol, chain, confidence/decimals, source timestamp and cache status. params: provider*, asset*, chain, force_refresh |
get_price_analytics | Run multi-oracle price analytics: consensus across providers, batch reads, historical series, cross-chain spreads, deviation, correlation or a risk summary. Use it when one price is not enough and the caller needs to compare, stress or trend oracle data. Do not use it for a single provider spot read (get_oracle_price) or for feed uptime (get_oracle_health). Returns the requested analysis rows plus params: assets*, mode, metric, providers, provider, method, chain, base_chain, from, to, interval, period_hours, force_refresh |
get_oracle_health | Read oracle ecosystem, feed, provider, protocol or asset health. Use it to answer freshness, uptime, latency, anomaly, coverage, peg, exposure, incident and daily-report questions before trusting a price or a protocol. Do not use it for a price quote (get_oracle_price) or for a trade gate (check_trade_safety). Returns the matching health rows, verdicts, counts and timestamps for the selected scope params: scope, subject, metric, period, days, date, from, to, category, is_active, min_severity, provider, chain_id, limit, offset |
search_oracle_registry | Search Insight-reviewed registry and evidence resources: symbols, feeds, protocols, stablecoins, protocol risk parameters, RWA instruments/context, settlement-evidence packs and oracle setup recommendations. Use it to discover the exact ids a price, safety or execution tool expects. Do not use it to read live prices or health. Returns matching registry rows or one lookup record, with ids and revie params: kind*, query, chain_id, category, is_active, limit, offset, verify_onchain |
get_market_quote | Read one reviewed venue market: pool/market metadata, pinned quote context and venue constraints for Base, Monad, Solana or Hyperliquid. Use it before preparing or gating a trade when the agent needs the reviewed profile or market id. Do not use it as execution authorization or for prices outside the reviewed venues. Returns the market record, reviewed profile ids and current quote context. params: venue*, profile_id, market, venues |
check_trade_safety | Run the pre-trade oracle safety gate for swaps, lending, perps, vault deposits and position stress tests across Base, Monad, multichain EVM, Solana and Hyperliquid. Use it immediately before an agent trades or signs, and pass the family native payload in `guard` when the venue needs exact-call fields. Do not use it after execution (use issue_execution_receipt) and do not treat its ALLOW as princip params: chain_id*, action*, asset, amount_usd, destination_asset, mode, protocol_id, market_id, market, profile_id, side, margin_asset, margin_value_usd, loan_symbol, collateral_value_usd, debt_value_usd, collateral_symbol, collateral_amount, borrow_symbol, borrow_amount, collaterals, borrows, position, providers, token_address, decision_input, guard, transaction, order, date, reputation_opt_in, reputation_agent_id |
begin_trade_evidence | Open the verifiable execution loop: certify a swap/DeFi intent before the agent executes. Use it when a principal needs a signed pre-trade handle to bind an execution receipt later. Do not use it as trade authorization and do not call it after the trade. Returns preTradeUid, destinationPreTradeUid, requestHash, CAIP-19 asset ids, certified cross-rate, the 50 bps band, verdicts and the signed pre-t params: asset*, destination_asset*, chain_id*, action*, amount_usd*, max_slippage_bps*, protocol_id, providers, source_group_count, decision_input* |
prepare_swap | Prepare unsigned, reviewed swap calldata for local signing on Base Aerodrome Slipstream, Base PancakeSwap V3, the Base agent swap guard, or a Solana Loopscale vault deposit. Use it when the user wants an executable transaction bound to a reviewed pool and policy. Do not use it to sign, submit or route arbitrary calldata, and do not skip the matching safety gate. Returns the unsigned transaction en params: venue*, direction, amount_in*, recipient, deadline_unix_seconds, max_slippage_bps, profile_id, source_decision_price_usd, destination_decision_price_usd, policy, wallet, vault, asset_mint, collateral_symbol, loan_symbol, collateral_value_usd, principal_amount, min_lp_amount |
prepare_perp_order | Prepare an unsigned Hyperliquid BTC/ETH/HYPE perpetual IOC order envelope for local wallet signing. Use it when an agent has decided on market, side, size and limit price and needs reviewed lot/tick, impact and mark-oracle checks. Do not use it to create a nonce, sign, or submit an order. Returns the short-expiry unsigned exchange envelope and the validated bounds when every gate passes. params: venue*, market*, side*, size*, limit_price*, reduce_only*, client_order_id*, max_notional_usd*, max_price_impact_bps*, max_mark_oracle_deviation_bps*, require_full_fill, expires_after_seconds, builder_fee |
issue_execution_receipt | Close the loop after an agent executes: issue a signed execution-price receipt from the pre-trade handle plus the settlement transaction, or settle a guarded Base/Solana swap. Use it only after the trade settled. Do not use it before execution and do not treat the verdict as principal authorization. Returns a signed receipt or settlement with FAITHFUL/DEVIATED/NOT_EXECUTED/UNDETERMINED (or settlem params: mode*, receipt, decision, swap_tx_hash, signature |
verify_evidence | Verify signed evidence produced by Insight or by an agent: a pre-trade/execution pair, a Base agent-swap settlement, a Solana swap settlement, or one Hyperliquid order/fill record. Use it when a third party needs to check binding and signatures before trusting a claim. Do not use it to re-assess oracle risk or to authorize trading. Returns valid/invalid, the closed-loop or settlement status, bindi params: kind*, pre_trade_attestation, execution_receipt, destination_pre_trade_attestation, settlement, user, order_id, submitted_at, expected_order |
oracle_watch | Read one oracle-trust snapshot or historical series for an asset. Use it for a one-off check of whether a symbol is currently trustworthy or how its recent history looks. Do not use it for continuous monitoring — buy a 24h window with oracle_watch_window_begin. Returns the current trust signal (or the requested history series) with provider participation and freshness. params: op, symbol*, chain, days, interval |
oracle_watch_window_begin | Buy a 24h Oracle Watch window for one asset and return the one-time control token. Use it when an agent will poll throughout a window instead of paying per snapshot. Do not use it repeatedly for the same window and never share the token. Returns windowId, the oww_ token, the covered expiry and the credits charged. params: symbol*, chain |
oracle_watch_window_control | Poll or close an already purchased Oracle Watch window using its one-time oww_ token. Poll returns a fresh signal at zero extra credits; close returns the complete window summary and the final signal. Use it only with a window token from oracle_watch_window_begin. Do not use it for a one-off snapshot (oracle_watch) and never reuse a token after close. Returns the signal (poll) or the combined summ params: op*, window_id*, window_token* |
position_lifecycle_begin | Buy a 24h Position Lifecycle bundle for one position subject and return the one-time control token. Use it when an agent needs guard, execution-receipt and close steps inside one prepaid window. Do not use it for a single guard (use check_trade_safety) and never share the token. Returns lifecycleId, the plc_ token, the covered expiry and the credits charged. params: symbol*, chain |
position_lifecycle_step | Run one step inside a purchased Position Lifecycle bundle: guard (family-specific safety assessment), receipt (signed execution receipt), or close (combined summary). Guard and receipt cost zero extra credits inside the bundle. Returns the step result (guard verdict, signed receipt, or the combined lifecycle summary). Use it only with a lifecycle token from position_lifecycle_begin. Do not use it params: op*, lifecycle_id*, lifecycle_token*, family, params, transaction, receipt |
Entries describe changes in tool definitions. They are not statements about a server, its quality or its author. The grade measures how the tool definitions fit model providers (token cost, schema compatibility, confusable names) as scored by the ToolFit engine at the time of the listing. JSON: append .json. Opt-out: how or hello@youraitools.co.